- 05 Jul, 2017 40 commits
-
-
Ladi Prosek authored
commit 6ed071f0 upstream. On AMD, the effect of set_nmi_mask called by emulate_iret_real and em_rsm on hflags is reverted later on in x86_emulate_instruction where hflags are overwritten with ctxt->emul_flags (the kvm_set_hflags call). This manifests as a hang when rebooting Windows VMs with QEMU, OVMF, and >1 vcpu. Instead of trying to merge ctxt->emul_flags into vcpu->arch.hflags after an instruction is emulated, this commit deletes emul_flags altogether and makes the emulator access vcpu->arch.hflags using two new accessors. This way all changes, on the emulator side as well as in functions called from the emulator and accessing vcpu state with emul_to_vcpu, are preserved. More details on the bug and its manifestation with Windows and OVMF: It's a KVM bug in the interaction between SMI/SMM and NMI, specific to AMD. I believe that the SMM part explains why we started seeing this only with OVMF. KVM masks and unmasks NMI when entering and leaving SMM. When KVM emulates the RSM instruction in em_rsm, the set_nmi_mask call doesn't stick because later on in x86_emulate_instruction we overwrite arch.hflags with ctxt->emul_flags, effectively reverting the effect of the set_nmi_mask call. The AMD-specific hflag of interest here is HF_NMI_MASK. When rebooting the system, Windows sends an NMI IPI to all but the current cpu to shut them down. Only after all of them are parked in HLT will the initiating cpu finish the restart. If NMI is masked, other cpus never get the memo and the initiating cpu spins forever, waiting for hal!HalpInterruptProcessorsStarted to drop. That's the symptom we observe. Fixes: a584539b ("KVM: x86: pass the whole hflags field to emulator and back") Signed-off-by:
Ladi Prosek <lprosek@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Thomas Petazzoni authored
commit ee56874f upstream. In commit eea62819 ("mtd: Add device-tree support to fsmc_nand"), Device Tree support was added to the fmsc_nand driver. However, this code has a bug in how it handles the bank-width DT property to set the bus width. Indeed, in the function fsmc_nand_probe_config_dt() that parses the Device Tree, it sets pdata->width to either 8 or 16 depending on the value of the bank-width DT property. Then, the ->probe() function will test if pdata->width is equal to FSMC_NAND_BW16 (which is 2) to set NAND_BUSWIDTH_16 in nand->options. Therefore, with the DT probing, this condition will never match. This commit fixes that by removing the "width" field from fsmc_nand_platform_data and instead have the fsmc_nand_probe_config_dt() function directly set the appropriate nand->options value. It is worth mentioning that if this commit gets backported to older kernels, prior to the drop of non-DT probing, then non-DT probing will be broken because nand->options will no longer be set to NAND_BUSWIDTH_16. Fixes: eea62819 ("mtd: Add device-tree support to fsmc_nand") Signed-off-by:
Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Reviewed-by:
Linus Walleij <linus.walleij@linaro.org> Signed-off-by:
Boris Brezillon <boris.brezillon@free-electrons.com> Signed-off-by:
-
Kamal Dasu authored
commit 9d2ee0a6 upstream. On brcmnand controller v6.x and v7.x, the #WP pin is controlled through the NAND_WP bit in CS_SELECT register. The driver currently assumes that toggling the #WP pin is instantaneously enabling/disabling write-protection, but it actually takes some time to propagate the new state to the internal NAND chip logic. This behavior is sometime causing data corruptions when an erase/program operation is executed before write-protection has really been disabled. Fixes: 27c5b17c ("mtd: nand: add NAND driver "library" for Broadcom STB NAND controller") Signed-off-by:
Kamal Dasu <kdasu.kdev@gmail.com> Signed-off-by:
-
Arnd Bergmann authored
commit 5b0ff9a0 upstream. hns_roce_v1_cq_set_ci() calls roce_set_bit() on an uninitialized field, which will then change only a few of its bits, causing a warning with the latest gcc: infiniband/hw/hns/hns_roce_hw_v1.c: In function 'hns_roce_v1_cq_set_ci': infiniband/hw/hns/hns_roce_hw_v1.c:1854:23: error: 'doorbell[1]' is used uninitialized in this function [-Werror=uninitialized] roce_set_bit(doorbell[1], ROCEE_DB_OTHERS_H_ROCEE_DB_OTH_HW_SYNS_S, 1); The code is actually correct since we always set all bits of the port_vlan field, but gcc correctly points out that the first access does contain uninitialized data. This initializes the field to zero first before setting the individual bits. Fixes: 9a443537 ("IB/hns: Add driver files for hns RoCE driver") Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
Doug Ledford <dledford@redhat.com> Signed-off-by:
-
Suravee Suthikulpanit authored
commit 84a21dbd upstream. Pass-through devices to VM guest can get updated IRQ affinity information via irq_set_affinity() when not running in guest mode. Currently, AMD IOMMU driver in GA mode ignores the updated information if the pass-through device is setup to use vAPIC regardless of guest_mode. This could cause invalid interrupt remapping. Also, the guest_mode bit should be set and cleared only when SVM updates posted-interrupt interrupt remapping information. Signed-off-by:
Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Cc: Joerg Roedel <jroedel@suse.de> Fixes: d98de49a ('iommu/amd: Enable vAPIC interrupt remapping mode by default') Signed-off-by:
Joerg Roedel <jroedel@suse.de> Signed-off-by:
-
Pan Bian authored
commit 73dbd4a4 upstream. In function amd_iommu_bind_pasid(), the control flow jumps to label out_free when pasid_state->mm and mm is NULL. And mmput(mm) is called. In function mmput(mm), mm is referenced without validation. This will result in a NULL dereference bug. This patch fixes the bug. Signed-off-by:
Pan Bian <bianpan2016@163.com> Fixes: f0aac63b ('iommu/amd: Don't hold a reference to mm_struct') Signed-off-by:
-
Robin Murphy authored
commit 938f1bbe upstream. Even if a host controller's CPU-side MMIO windows into PCI I/O space do happen to leak into PCI memory space such that it might treat them as peer addresses, trying to reserve the corresponding I/O space addresses doesn't do anything to help solve that problem. Stop doing a silly thing. Fixes: fade1ec0 ("iommu/dma: Avoid PCI host bridge windows") Reviewed-by:
Eric Auger <eric.auger@redhat.com> Signed-off-by:
Robin Murphy <robin.murphy@arm.com> Signed-off-by:
-
Eric Ren authored
commit 8818efaa upstream. Another deadlock path caused by recursive locking is reported. This kind of issue was introduced since commit 743b5f14 ("ocfs2: take inode lock in ocfs2_iop_set/get_acl()"). Two deadlock paths have been fixed by commit b891fa50 ("ocfs2: fix deadlock issue when taking inode lock at vfs entry points"). Yes, we intend to fix this kind of case in incremental way, because it's hard to find out all possible paths at once. This one can be reproduced like this. On node1, cp a large file from home directory to ocfs2 mountpoint. While on node2, run setfacl/getfacl. Both nodes will hang up there. The backtraces: On node1: __ocfs2_cluster_lock.isra.39+0x357/0x740 [ocfs2] ocfs2_inode_lock_full_nested+0x17d/0x840 [ocfs2] ocfs2_write_begin+0x43/0x1a0 [ocfs2] generic_perform_write+0xa9/0x180 __generic_file_write_iter+0x1aa/0x1d0 ocfs2_file_write_iter+0x4f4/0xb40 [ocfs2] __vfs_write+0xc3/0x130 vfs_write+0xb1/0x1a0 SyS_write+0x46/0xa0 On node2: __ocfs2_cluster_lock.isra.39+0x357/0x740 [ocfs2] ocfs2_inode_lock_full_nested+0x17d/0x840 [ocfs2] ocfs2_xattr_set+0x12e/0xe80 [ocfs2] ocfs2_set_acl+0x22d/0x260 [ocfs2] ocfs2_iop_set_acl+0x65/0xb0 [ocfs2] set_posix_acl+0x75/0xb0 posix_acl_xattr_set+0x49/0xa0 __vfs_setxattr+0x69/0x80 __vfs_setxattr_noperm+0x72/0x1a0 vfs_setxattr+0xa7/0xb0 setxattr+0x12d/0x190 path_setxattr+0x9f/0xb0 SyS_setxattr+0x14/0x20 Fix this one by using ocfs2_inode_{lock|unlock}_tracker, which is exported by commit 439a36b8 ("ocfs2/dlmglue: prepare tracking logic to avoid recursive cluster lock"). Link: http://lkml.kernel.org/r/20170622014746.5815-1-zren@suse.com Fixes: 743b5f14 ("ocfs2: take inode lock in ocfs2_iop_set/get_acl()") Signed-off-by:
Eric Ren <zren@suse.com> Reported-by:
Thomas Voegtle <tv@lio96.de> Tested-by:
Joseph Qi <jiangqi903@gmail.com> Cc: Mark Fasheh <mfasheh@versity.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
-
Junxiao Bi authored
commit 33496c3c upstream. Configfs is the interface for ocfs2-tools to set configure to kernel and $configfs_dir/cluster/$clustername/heartbeat/dead_threshold is the one used to configure heartbeat dead threshold. Kernel has a default value of it but user can set O2CB_HEARTBEAT_THRESHOLD in /etc/sysconfig/o2cb to override it. Commit 45b99773 ("ocfs2/cluster: use per-attribute show and store methods") changed heartbeat dead threshold name while ocfs2-tools did not, so ocfs2-tools won't set this configurable and the default value is always used. So revert it. Fixes: 45b99773 ("ocfs2/cluster: use per-attribute show and store methods") Link: http://lkml.kernel.org/r/1490665245-15374-1-git-send-email-junxiao.bi@oracle.comSigned-off-by:
Junxiao Bi <junxiao.bi@oracle.com> Acked-by:
-
Andy Lutomirski authored
commit dbd68d8e upstream. flush_tlb_page() passes a bogus range to flush_tlb_others() and expects the latter to fix it up. native_flush_tlb_others() has the fixup but Xen's version doesn't. Move the fixup to flush_tlb_others(). AFAICS the only real effect is that, without this fix, Xen would flush everything instead of just the one page on remote vCPUs in when flush_tlb_page() was called. Signed-off-by:
Andy Lutomirski <luto@kernel.org> Reviewed-by:
Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Juergen Gross <jgross@suse.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Nadav Amit <namit@vmware.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rik van Riel <riel@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Fixes: e7b52ffd ("x86/flush_tlb: try flush_tlb_single one by one in flush_tlb_range") Link: http://lkml.kernel.org/r/10ed0e4dfea64daef10b87fb85df1746999b4dba.1492844372.git.luto@kernel.orgSigned-off-by:
Ingo Molnar <mingo@kernel.org> Signed-off-by:
-
Joerg Roedel authored
commit 5ed386ec upstream. When this function fails it just sends a SIGSEGV signal to user-space using force_sig(). This signal is missing essential information about the cause, e.g. the trap_nr or an error code. Fix this by propagating the error to the only caller of mpx_handle_bd_fault(), do_bounds(), which sends the correct SIGSEGV signal to the process. Signed-off-by:
-
Kan Liang authored
commit fd583ad1 upstream. Spurious NMIs will be observed with the following command: while :; do perf record -bae "cpu/umask=0x01,event=0xcd,ldlat=0x80/pp" -e "cpu/umask=0x03,event=0x0/" -e "cpu/umask=0x02,event=0x0/" -e cycles,branches,cache-misses -e cache-references -- sleep 10 done The bug was introduced by commit: 8077eca0 ("perf/x86/pebs: Add workaround for broken OVFL status on HSW+") That commit clears the status bits for the counters used for PEBS events, by masking the whole 64 bits pebs_enabled. However, only the low 32 bits of both status and pebs_enabled are reserved for PEBS-able counters. For status bits 32-34 are fixed counter overflow bits. For pebs_enabled bits 32-34 are for PEBS Load Latency. In the test case, the PEBS Load Latency event and fixed counter event could overflow at the same time. The fixed counter overflow bit will be cleared by mistake. Once it is cleared, the fixed counter overflow never be processed, which finally trigger spurious NMI. Correct the PEBS enabled mask by ignoring the non-PEBS bits. Signed-off-by:
Kan Liang <kan.liang@intel.com> Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vince Weaver <vincent.weaver@maine.edu> Fixes: 8077eca0 ("perf/x86/pebs: Add workaround for broken OVFL status on HSW+") Link: http://lkml.kernel.org/r/1491333246-3965-1-git-send-email-kan.liang@intel.comSigned-off-by:
-
Baoquan He authored
commit 8eabf42a upstream. Kernel text KASLR is separated into physical address and virtual address randomization. And for virtual address randomization, we only randomiza to get an offset between 16M and KERNEL_IMAGE_SIZE. So the initial value of 'virt_addr' should be LOAD_PHYSICAL_ADDR, but not the original kernel loading address 'output'. The bug will cause kernel boot failure if kernel is loaded at a different position than the address, 16M, which is decided at compiled time. Kexec/kdump is such practical case. To fix it, just assign LOAD_PHYSICAL_ADDR to virt_addr as initial value. Tested-by:
Dave Young <dyoung@redhat.com> Signed-off-by:
Baoquan He <bhe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Fixes: 8391c73c ("x86/KASLR: Randomize virtual address separately") Link: http://lkml.kernel.org/r/1498567146-11990-3-git-send-email-bhe@redhat.comSigned-off-by:
-
Thomas Gleixner authored
commit 26fcd952 upstream. A recent commit included linux/slab.h in linux/irq.h. This breaks the build of vdso32 on a 64-bit kernel. The reason is that linux/irq.h gets included into the vdso code via linux/interrupt.h which is included from asm/mshyperv.h. That makes the 32-bit vdso compile fail, because slab.h includes the pgtable headers for 64-bit on a 64-bit build. Neither linux/clocksource.h nor linux/interrupt.h are needed in the mshyperv.h header file itself - it has a dependency on <linux/atomic.h>. Remove the includes and unbreak the build. Reported-by:
Thomas Gleixner <tglx@linutronix.de> Cc: K. Y. Srinivasan <kys@microsoft.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Vitaly Kuznetsov <vkuznets@redhat.com> Cc: devel@linuxdriverproject.org Fixes: dee863b5 ("hv: export current Hyper-V clocksource") Link: http://lkml.kernel.org/r/alpine.DEB.2.20.1706231038460.2647@nanosSigned-off-by:
-
Josh Poimboeuf authored
commit ebd57499 upstream. Petr Mladek reported the following warning when loading the livepatch sample module: WARNING: CPU: 1 PID: 3699 at arch/x86/kernel/stacktrace.c:132 save_stack_trace_tsk_reliable+0x133/0x1a0 ... Call Trace: __schedule+0x273/0x820 schedule+0x36/0x80 kthreadd+0x305/0x310 ? kthread_create_on_cpu+0x80/0x80 ? icmp_echo.part.32+0x50/0x50 ret_from_fork+0x2c/0x40 That warning means the end of the stack is no longer recognized as such for newly forked tasks. The problem was introduced with the following commit: ff3f7e24 ("x86/entry: Fix the end of the stack for newly forked tasks") ... which was completely misguided. It only partially fixed the reported issue, and it introduced another bug in the process. None of the other entry code saves the frame pointer before calling into C code, so it doesn't make sense for ret_from_fork to do so either. Contrary to what I originally thought, the original issue wasn't related to newly forked tasks. It was actually related to ftrace. When entry code calls into a function which then calls into an ftrace handler, the stack frame looks different than normal. The original issue will be fixed in the unwinder, in a subsequent patch. Reported-by:
Petr Mladek <pmladek@suse.com> Signed-off-by:
Josh Poimboeuf <jpoimboe@redhat.com> Acked-by:
-
Arnaldo Carvalho de Melo authored
commit e883d09c upstream. Just a minor fix done in: Fixes: 26a37ab3 ("x86/mce: Fix copy/paste error in exception table entries") Cc: Tony Luck <tony.luck@intel.com> Link: http://lkml.kernel.org/n/tip-ni9jzdd5yxlail6pq8cuexw2@git.kernel.orgSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by:
-
Christophe JAILLET authored
commit 95d7c1f1 upstream. It is wrong to iounmap resources in the normal path of davinci_pm_init() The 3 ioremap'ed fields of 'pm_config' can be accessed later on in other functions, so we should return 'success' instead of unrolling everything. Fixes: aa9aa1ec ("ARM: davinci: PM: rework init, remove platform device") Signed-off-by:
Christophe JAILLET <christophe.jaillet@wanadoo.fr> [nsekhar@ti.com: commit message and minor style fixes] Signed-off-by:
Sekhar Nori <nsekhar@ti.com> Signed-off-by:
-
Christophe JAILLET authored
commit f3f6cc81 upstream. If 'sram_alloc' fails, we need to free already allocated resources. Fixes: aa9aa1ec ("ARM: davinci: PM: rework init, remove platform device") Signed-off-by:
-
Doug Berger authored
commit 9e25ebfe upstream. The pmd containing memblock_limit is cleared by prepare_page_table() which creates the opportunity for early_alloc() to allocate unmapped memory if memblock_limit is not pmd aligned causing a boot-time hang. Commit 965278dc ("ARM: 8356/1: mm: handle non-pmd-aligned end of RAM") attempted to resolve this problem, but there is a path through the adjust_lowmem_bounds() routine where if all memory regions start and end on pmd-aligned addresses the memblock_limit will be set to arm_lowmem_limit. Since arm_lowmem_limit can be affected by the vmalloc early parameter, the value of arm_lowmem_limit may not be pmd-aligned. This commit corrects this oversight such that memblock_limit is always rounded down to pmd-alignment. Fixes: 965278dc ("ARM: 8356/1: mm: handle non-pmd-aligned end of RAM") Signed-off-by:
Doug Berger <opendmb@gmail.com> Suggested-by:
Mark Rutland <mark.rutland@arm.com> Signed-off-by:
Russell King <rmk+kernel@armlinux.org.uk> Signed-off-by:
-
Lorenzo Pieralisi authored
commit cb7cf772 upstream. The BAD_MADT_GICC_ENTRY() macro checks if a GICC MADT entry passes muster from an ACPI specification standpoint. Current macro detects the MADT GICC entry length through ACPI firmware version (it changed from 76 to 80 bytes in the transition from ACPI 5.1 to ACPI 6.0 specification) but always uses (erroneously) the ACPICA (latest) struct (ie struct acpi_madt_generic_interrupt - that is 80-bytes long) length to check if the current GICC entry memory record exceeds the MADT table end in memory as defined by the MADT table header itself, which may result in false negatives depending on the ACPI firmware version and how the MADT entries are laid out in memory (ie on ACPI 5.1 firmware MADT GICC entries are 76 bytes long, so by adding 80 to a GICC entry start address in memory the resulting address may well be past the actual MADT end, triggering a false negative). Fix the BAD_MADT_GICC_ENTRY() macro by reshuffling the condition checks and update them to always use the firmware version specific MADT GICC entry length in order to carry out boundary checks. Fixes: b6cfb277 ("ACPI / ARM64: add BAD_MADT_GICC_ENTRY() macro") Reported-by:
Julien Grall <julien.grall@arm.com> Acked-by:
Will Deacon <will.deacon@arm.com> Acked-by:
Marc Zyngier <marc.zyngier@arm.com> Signed-off-by:
Lorenzo Pieralisi <lorenzo.pieralisi@arm.com> Cc: Julien Grall <julien.grall@arm.com> Cc: Hanjun Guo <hanjun.guo@linaro.org> Cc: Al Stone <ahs3@redhat.com> Signed-off-by:
Catalin Marinas <catalin.marinas@arm.com> Signed-off-by:
-
Timmy Li authored
commit 717902cc upstream. Commit 093d24a2 ("arm64: PCI: Manage controller-specific data on per-controller basis") added code to allocate ACPI PCI root_ops dynamically on a per host bridge basis but failed to update the corresponding memory allocation failure path in pci_acpi_scan_root() leading to a potential memory leakage. Fix it by adding the required kfree call. Fixes: 093d24a2 ("arm64: PCI: Manage controller-specific data on per-controller basis") Reviewed-by:
Tomasz Nowicki <tn@semihalf.com> Signed-off-by:
Timmy Li <lixiaoping3@huawei.com> [lorenzo.pieralisi@arm.com: refactored code, rewrote commit log] Signed-off-by:
-
Eric Anholt authored
commit fedf266f upstream. The bcm_kona_wdt_set_resolution_reg() call takes the spinlock, so initialize it earlier. Fixes a warning at boot with lock debugging enabled. Fixes: 6adb730d ("watchdog: bcm281xx: Watchdog Driver") Signed-off-by:
Eric Anholt <eric@anholt.net> Reviewed-by:
Florian Fainelli <f.fainelli@gmail.com> Reviewed-by:
Guenter Roeck <linux@roeck-us.net> Signed-off-by:
Wim Van Sebroeck <wim@iguana.be> Signed-off-by:
-
Dan Carpenter authored
commit 1e3d0c2c upstream. There are some missing error codes here so we accidentally return NULL instead of an error pointer. It results in a NULL pointer dereference. Fixes: df71837d ("[LSM-IPSec]: Security association restriction.") Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by:
-
Dan Carpenter authored
commit e747f643 upstream. The default error code in pfkey_msg2xfrm_state() is -ENOBUFS. We added a new call to security_xfrm_state_alloc() which sets "err" to zero so there several places where we can return ERR_PTR(0) if kmalloc() fails. The caller is expecting error pointers so it leads to a NULL dereference. Fixes: df71837d ("[LSM-IPSec]: Security association restriction.") Signed-off-by:
-
Sabrina Dubroca authored
commit 9b3eb541 upstream. When CONFIG_XFRM_SUB_POLICY=y, xfrm_dst stores a copy of the flowi for that dst. Unfortunately, the code that allocates and fills this copy doesn't care about what type of flowi (flowi, flowi4, flowi6) gets passed. In multiple code paths (from raw_sendmsg, from TCP when replying to a FIN, in vxlan, geneve, and gre), the flowi that gets passed to xfrm is actually an on-stack flowi4, so we end up reading stuff from the stack past the end of the flowi4 struct. Since xfrm_dst->origin isn't used anywhere following commit ca116922 ("xfrm: Eliminate "fl" and "pol" args to xfrm_bundle_ok()."), just get rid of it. xfrm_dst->partner isn't used either, so get rid of that too. Fixes: 9d6ec938 ("ipv4: Use flowi4 in public route lookup interfaces.") Signed-off-by:
Sabrina Dubroca <sd@queasysnail.net> Signed-off-by:
-
Hangbin Liu authored
commit 138437f5 upstream. Now we will force to do garbage collection if any policy removed in xfrm_policy_flush(). But during xfrm_net_exit(). We call flow_cache_fini() first and set set fc->percpu to NULL. Then after we call xfrm_policy_fini() -> frxm_policy_flush() -> flow_cache_flush(), we will get NULL pointer dereference when check percpu_empty. The code path looks like: flow_cache_fini() - fc->percpu = NULL xfrm_policy_fini() - xfrm_policy_flush() - xfrm_garbage_collect() - flow_cache_flush() - flow_cache_percpu_empty() - fcp = per_cpu_ptr(fc->percpu, cpu) To reproduce, just add ipsec in netns and then remove the netns. v2: As Xin Long suggested, since only two other places need to call it. move xfrm_garbage_collect() outside xfrm_policy_flush(). v3: Fix subject mismatch after v2 fix. Fixes: 35db0691 ("xfrm: do the garbage collection after flushing policy") Signed-off-by:
Hangbin Liu <liuhangbin@gmail.com> Reviewed-by:
Xin Long <lucien.xin@gmail.com> Signed-off-by:
-
Yossi Kuperman authored
commit 7c88e21a upstream. IPv6 payload length indicates the size of the payload, including any extension headers. In xfrm6_transport_finish, ipv6_hdr(skb)->payload_len is set to the payload size only, regardless of the presence of any extension headers. After ESP GRO transport mode decapsulation, ipv6_rcv trims the packet according to the wrong payload_len, thus corrupting the packet. Set payload_len to account for extension headers as well. Fixes: 7785bba2 ("esp: Add a software GRO codepath") Signed-off-by:
Yossi Kuperman <yossiku@mellanox.com> Signed-off-by:
-
Juergen Gross authored
commit 71df1d7c upstream. The be structure must not be freed when freeing the blkif structure isn't done. Otherwise a use-after-free of be when unmapping the ring used for communicating with the frontend will occur in case of a late call of xenblk_disconnect() (e.g. due to an I/O still active when trying to disconnect). Signed-off-by:
Juergen Gross <jgross@suse.com> Tested-by:
Steven Haigh <netwiz@crc.id.au> Acked-by:
Roger Pau Monné <roger.pau@citrix.com> Signed-off-by:
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by:
-
Ard Biesheuvel authored
commit 029c54b0 upstream. Existing code that uses vmalloc_to_page() may assume that any address for which is_vmalloc_addr() returns true may be passed into vmalloc_to_page() to retrieve the associated struct page. This is not un unreasonable assumption to make, but on architectures that have CONFIG_HAVE_ARCH_HUGE_VMAP=y, it no longer holds, and we need to ensure that vmalloc_to_page() does not go off into the weeds trying to dereference huge PUDs or PMDs as table entries. Given that vmalloc() and vmap() themselves never create huge mappings or deal with compound pages at all, there is no correct answer in this case, so return NULL instead, and issue a warning. When reading /proc/kcore on arm64, you will hit an oops as soon as you hit the huge mappings used for the various segments that make up the mapping of vmlinux. With this patch applied, you will no longer hit the oops, but the kcore contents willl be incorrect (these regions will be zeroed out) We are fixing this for kcore specifically, so it avoids vread() for those regions. At least one other problematic user exists, i.e., /dev/kmem, but that is currently broken on arm64 for other reasons. Link: http://lkml.kernel.org/r/20170609082226.26152-1-ard.biesheuvel@linaro.orgSigned-off-by:
Ard Biesheuvel <ard.biesheuvel@linaro.org> Acked-by:
Laura Abbott <labbott@redhat.com> Cc: Michal Hocko <mhocko@suse.com> Cc: zhong jiang <zhongjiang@huawei.com> Cc: Dave Hansen <dave.hansen@intel.com> Signed-off-by:
-
Thomas Gleixner authored
commit ba714a9c upstream. The AMD pinctrl driver uses a chained interrupt to demultiplex the GPIO interrupts. Kevin Vandeventer reported, that his new AMD Ryzen locks up hard on boot when the AMD pinctrl driver is initialized. The reason is an interrupt storm. It's not clear whether that's caused by hardware or firmware or both. Using chained interrupts on X86 is a dangerous endavour. If a system is misconfigured or the hardware buggy there is no safety net to catch an interrupt storm. Convert the driver to use a regular interrupt for the demultiplex handler. This allows the interrupt storm detector to catch the malfunction and lets the system boot up. This should be backported to stable because it's likely that more users run into this problem as the AMD Ryzen machines are spreading. Reported-by: Kevin Vandeventer Link: https://bugzilla.suse.com/show_bug.cgi?id=1034261Signed-off-by:
Borislav Petkov <bp@suse.de> Signed-off-by:
-
Baoquan He authored
commit fc5f9d5f upstream. Jeff Moyer reported that on his system with two memory regions 0~64G and 1T~1T+192G, and kernel option "memmap=192G!1024G" added, enabling KASLR will make the system hang intermittently during boot. While adding 'nokaslr' won't. The back trace is: Oops: 0000 [#1] SMP RIP: memcpy_erms() [ .... ] Call Trace: pmem_rw_page() bdev_read_page() do_mpage_readpage() mpage_readpages() blkdev_readpages() __do_page_cache_readahead() force_page_cache_readahead() page_cache_sync_readahead() generic_file_read_iter() blkdev_read_iter() __vfs_read() vfs_read() SyS_read() entry_SYSCALL_64_fastpath() This crash happens because the for loop count calculation in sync_global_pgds() is not correct. When a mapping area crosses PGD entries, we should calculate the starting address of region which next PGD covers and assign it to next for loop count, but not add PGDIR_SIZE directly. The old code works right only if the mapping area is an exact multiple of PGDIR_SIZE, otherwize the end region could be skipped so that it can't be synchronized to all other processes from kernel PGD init_mm.pgd. In Jeff's system, emulated pmem area [1024G, 1216G) is smaller than PGDIR_SIZE. While 'nokaslr' works because PAGE_OFFSET is 1T aligned, it makes this area be mapped inside one PGD entry. With KASLR enabled, this area could cross two PGD entries, then the next PGD entry won't be synced to all other processes. That is why we saw empty PGD. Fix it. Reported-by:
Jeff Moyer <jmoyer@redhat.com> Signed-off-by:
Dan Williams <dan.j.williams@intel.com> Signed-off-by:
-
Vallish Vaidyeshwara authored
commit 00a0ea33 upstream. process_prepared_discard_passdown_pt1() should cleanup dm_thin_new_mapping in cases of error. dm_pool_inc_data_range() can fail trying to get a block reference: metadata operation 'dm_pool_inc_data_range' failed: error = -61 When dm_pool_inc_data_range() fails, dm thin aborts current metadata transaction and marks pool as PM_READ_ONLY. Memory for thin mapping is released as well. However, current thin mapping will be queued onto next stage as part of queue_passdown_pt2() or passdown_endio(). This dangling thin mapping memory when processed and accessed in next stage will lead to device mapper crashing. Code flow without fix: -> process_prepared_discard_passdown_pt1(m) -> dm_thin_remove_range() -> discard passdown --> passdown_endio(m) queues m onto next stage -> dm_pool_inc_data_range() fails, frees memory m but does not remove it from next stage queue -> process_prepared_discard_passdown_pt2(m) -> processes freed memory m and crashes One such stack: Call Trace: [<ffffffffa037a46f>] dm_cell_release_no_holder+0x2f/0x70 [dm_bio_prison] [<ffffffffa039b6dc>] cell_defer_no_holder+0x3c/0x80 [dm_thin_pool] [<ffffffffa039b88b>] process_prepared_discard_passdown_pt2+0x4b/0x90 [dm_thin_pool] [<ffffffffa0399611>] process_prepared+0x81/0xa0 [dm_thin_pool] [<ffffffffa039e735>] do_worker+0xc5/0x820 [dm_thin_pool] [<ffffffff8152bf54>] ? __schedule+0x244/0x680 [<ffffffff81087e72>] ? pwq_activate_delayed_work+0x42/0xb0 [<ffffffff81089f53>] process_one_work+0x153/0x3f0 [<ffffffff8108a71b>] worker_thread+0x12b/0x4b0 [<ffffffff8108a5f0>] ? rescuer_thread+0x350/0x350 [<ffffffff8108fd6a>] kthread+0xca/0xe0 [<ffffffff8108fca0>] ? kthread_park+0x60/0x60 [<ffffffff81530b45>] ret_from_fork+0x25/0x30 The fix is to first take the block ref count for discarded block and then do a passdown discard of this block. If block ref count fails, then bail out aborting current metadata transaction, mark pool as PM_READ_ONLY and also free current thin mapping memory (existing error handling code) without queueing this thin mapping onto next stage of processing. If block ref count succeeds, then passdown discard of this block. Discard callback of passdown_endio() will queue this thin mapping onto next stage of processing. Code flow with fix: -> process_prepared_discard_passdown_pt1(m) -> dm_thin_remove_range() -> dm_pool_inc_data_range() --> if fails, free memory m and bail out -> discard passdown --> passdown_endio(m) queues m onto next stage Reviewed-by:
Eduardo Valentin <eduval@amazon.com> Reviewed-by:
Cristian Gafton <gafton@amazon.com> Reviewed-by:
Anchal Agarwal <anchalag@amazon.com> Signed-off-by:
Vallish Vaidyeshwara <vallish@amazon.com> Reviewed-by:
Joe Thornber <ejt@redhat.com> Signed-off-by:
Mike Snitzer <snitzer@redhat.com> Signed-off-by:
-
Deepak Rawat authored
commit 82fcee52 upstream. The hash table created during vmw_cmdbuf_res_man_create was never freed. This causes memory leak in context creation. Added the corresponding drm_ht_remove in vmw_cmdbuf_res_man_destroy. Tested for memory leak by running piglit overnight and kernel memory is not inflated which earlier was. Signed-off-by:
Deepak Rawat <drawat@vmware.com> Reviewed-by:
Sinclair Yeh <syeh@vmware.com> Signed-off-by:
Thomas Hellstrom <thellstrom@vmware.com> Signed-off-by:
-
Kan Liang authored
commit 80c65fdb upstream. Should not init a NULL box. It will cause system crash. The issue looks like caused by a typo. This was not noticed because there is no NULL box. Also, for most boxes, they are enabled by default. The init code is not critical. Fixes: fff4b87e ("perf/x86/intel/uncore: Make package handling more robust") Signed-off-by:
-
Vikas Shivappa authored
commit 79298acc upstream. If mount fails, the kn_info directory is not freed causing memory leak. Add the missing error handling path. Fixes: 4e978d06 ("x86/intel_rdt: Add "info" files to resctrl file system") Signed-off-by:
Vikas Shivappa <vikas.shivappa@linux.intel.com> Signed-off-by:
-
Bartosz Golaszewski authored
commit ad537b82 upstream. GPIOEVENT_REQUEST_BOTH_EDGES is not a single flag, but a binary OR of GPIOEVENT_REQUEST_RISING_EDGE and GPIOEVENT_REQUEST_FALLING_EDGE. The expression 'le->eflags & GPIOEVENT_REQUEST_BOTH_EDGES' we'll get evaluated to true even if only one event type was requested. Fix it by checking both RISING & FALLING flags explicitly. Fixes: 61f922db ("gpio: userspace ABI for reading GPIO line events") Signed-off-by:
Bartosz Golaszewski <brgl@bgdev.pl> Signed-off-by:
-
Miklos Szeredi authored
commit e85f82ff upstream. Nothing prevents mischief on upper layer while we are busy copying up the data. Move the lookup right before the looked up dentry is actually used. Signed-off-by:
Miklos Szeredi <mszeredi@redhat.com> Fixes: 01ad3eb8 ("ovl: concurrent copy up of regular files") Signed-off-by:
-
Benjamin Coddington authored
commit d9f29500 upstream. This reverts commit 920b4530 which could call d_move() without holding the directory's i_mutex, and reverts commit d4ea7e3c "NFS: Fix old dentry rehash after move", which was a follow-up fix. Signed-off-by:
Benjamin Coddington <bcodding@redhat.com> Fixes: 920b4530 ("NFS: nfs_rename() handle -ERESTARTSYS dentry left behind") Reviewed-by:
Jeff Layton <jlayton@redhat.com> Signed-off-by:
Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by:
-
Trond Myklebust authored
commit bd171930 upstream. If the task calling layoutget is signalled, then it is possible for the calls to nfs4_sequence_free_slot() and nfs4_layoutget_prepare() to race, in which case we leak a slot. The fix is to move the call to nfs4_sequence_free_slot() into the nfs4_layoutget_release() so that it gets called at task teardown time. Fixes: 2e80dbe7 ("NFSv4.1: Close callback races for OPEN, LAYOUTGET...") Signed-off-by:
-
Benjamin Coddington authored
commit 501e7a46 upstream. Now that we have umask support, we shouldn't re-send the mode in a SETATTR following an exclusive CREATE, or we risk having the same problem fixed in commit 5334c5bd ("NFS: Send attributes in OPEN request for NFS4_CREATE_EXCLUSIVE4_1"), which is that files with S_ISGID will have that bit stripped away. Signed-off-by:
-
