platform/kernel/linux-starfive.git
2 years agox86/kvm: Don't waste memory if kvmclock is disabled
Wanpeng Li [Tue, 22 Feb 2022 09:02:03 +0000 (01:02 -0800)]
x86/kvm: Don't waste memory if kvmclock is disabled

Even if "no-kvmclock" is passed in cmdline parameter, the guest kernel
still allocates hvclock_mem which is scaled by the number of vCPUs,
let's check kvmclock enable in advance to avoid this memory waste.

Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Message-Id: <1645520523-30814-1-git-send-email-wanpengli@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agox86/kvm: Don't use PV TLB/yield when mwait is advertised
Wanpeng Li [Fri, 25 Feb 2022 08:29:40 +0000 (00:29 -0800)]
x86/kvm: Don't use PV TLB/yield when mwait is advertised

MWAIT is advertised in host is not overcommitted scenario, however, PV
TLB/sched yield should be enabled in host overcommitted scenario. Let's
add the MWAIT checking when enabling PV TLB/sched yield.

Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Message-Id: <1645777780-2581-1-git-send-email-wanpengli@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoMerge tag 'kvmarm-fixes-5.17-4' of git://git.kernel.org/pub/scm/linux/kernel/git...
Paolo Bonzini [Fri, 25 Feb 2022 14:49:30 +0000 (09:49 -0500)]
Merge tag 'kvmarm-fixes-5.17-4' of git://git./linux/kernel/git/kvmarm/kvmarm into HEAD

KVM/arm64 fixes for 5.17, take #4

- Correctly synchronise PMR and co on PSCI CPU_SUSPEND

- Skip tests that depend on GICv3 when the HW isn't available

2 years agoKVM: selftests: aarch64: Skip tests if we can't create a vgic-v3
Mark Brown [Wed, 23 Feb 2022 13:16:24 +0000 (13:16 +0000)]
KVM: selftests: aarch64: Skip tests if we can't create a vgic-v3

The arch_timer and vgic_irq kselftests assume that they can create a
vgic-v3, using the library function vgic_v3_setup() which aborts with a
test failure if it is not possible to do so. Since vgic-v3 can only be
instantiated on systems where the host has GICv3 this leads to false
positives on older systems where that is not the case.

Fix this by changing vgic_v3_setup() to return an error if the vgic can't
be instantiated and have the callers skip if this happens. We could also
exit flagging a skip in vgic_v3_setup() but this would prevent future test
cases conditionally deciding which GIC to use or generally doing more
complex output.

Signed-off-by: Mark Brown <broonie@kernel.org>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Tested-by: Ricardo Koller <ricarkol@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220223131624.1830351-1-broonie@kernel.org
2 years agoRevert "KVM: VMX: Save HOST_CR3 in vmx_prepare_switch_to_guest()"
Sean Christopherson [Thu, 24 Feb 2022 19:19:17 +0000 (19:19 +0000)]
Revert "KVM: VMX: Save HOST_CR3 in vmx_prepare_switch_to_guest()"

Revert back to refreshing vmcs.HOST_CR3 immediately prior to VM-Enter.
The PCID (ASID) part of CR3 can be bumped without KVM being scheduled
out, as the kernel will switch CR3 during __text_poke(), e.g. in response
to a static key toggling.  If switch_mm_irqs_off() chooses a new ASID for
the mm associate with KVM, KVM will do VM-Enter => VM-Exit with a stale
vmcs.HOST_CR3.

Add a comment to explain why KVM must wait until VM-Enter is imminent to
refresh vmcs.HOST_CR3.

The following splat was captured by stashing vmcs.HOST_CR3 in kvm_vcpu
and adding a WARN in load_new_mm_cr3() to fire if a new ASID is being
loaded for the KVM-associated mm while KVM has a "running" vCPU:

  static void load_new_mm_cr3(pgd_t *pgdir, u16 new_asid, bool need_flush)
  {
struct kvm_vcpu *vcpu = kvm_get_running_vcpu();

...

WARN(vcpu && (vcpu->cr3 & GENMASK(11, 0)) != (new_mm_cr3 & GENMASK(11, 0)) &&
     (vcpu->cr3 & PHYSICAL_PAGE_MASK) == (new_mm_cr3 & PHYSICAL_PAGE_MASK),
     "KVM is hosed, loading CR3 = %lx, vmcs.HOST_CR3 = %lx", new_mm_cr3, vcpu->cr3);
  }

  ------------[ cut here ]------------
  KVM is hosed, loading CR3 = 8000000105393004, vmcs.HOST_CR3 = 105393003
  WARNING: CPU: 4 PID: 20717 at arch/x86/mm/tlb.c:291 load_new_mm_cr3+0x82/0xe0
  Modules linked in: vhost_net vhost vhost_iotlb tap kvm_intel
  CPU: 4 PID: 20717 Comm: stable Tainted: G        W         5.17.0-rc3+ #747
  Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015
  RIP: 0010:load_new_mm_cr3+0x82/0xe0
  RSP: 0018:ffffc9000489fa98 EFLAGS: 00010082
  RAX: 0000000000000000 RBX: 8000000105393004 RCX: 0000000000000027
  RDX: 0000000000000027 RSI: 00000000ffffdfff RDI: ffff888277d1b788
  RBP: 0000000000000004 R08: ffff888277d1b780 R09: ffffc9000489f8b8
  R10: 0000000000000001 R11: 0000000000000001 R12: 0000000000000000
  R13: ffff88810678a800 R14: 0000000000000004 R15: 0000000000000c33
  FS:  00007fa9f0e72700(0000) GS:ffff888277d00000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000000 CR3: 00000001001b5003 CR4: 0000000000172ea0
  Call Trace:
   <TASK>
   switch_mm_irqs_off+0x1cb/0x460
   __text_poke+0x308/0x3e0
   text_poke_bp_batch+0x168/0x220
   text_poke_finish+0x1b/0x30
   arch_jump_label_transform_apply+0x18/0x30
   static_key_slow_inc_cpuslocked+0x7c/0x90
   static_key_slow_inc+0x16/0x20
   kvm_lapic_set_base+0x116/0x190
   kvm_set_apic_base+0xa5/0xe0
   kvm_set_msr_common+0x2f4/0xf60
   vmx_set_msr+0x355/0xe70 [kvm_intel]
   kvm_set_msr_ignored_check+0x91/0x230
   kvm_emulate_wrmsr+0x36/0x120
   vmx_handle_exit+0x609/0x6c0 [kvm_intel]
   kvm_arch_vcpu_ioctl_run+0x146f/0x1b80
   kvm_vcpu_ioctl+0x279/0x690
   __x64_sys_ioctl+0x83/0xb0
   do_syscall_64+0x3b/0xc0
   entry_SYSCALL_64_after_hwframe+0x44/0xae
   </TASK>
  ---[ end trace 0000000000000000 ]---

This reverts commit 15ad9762d69fd8e40a4a51828c1d6b0c1b8fbea0.

Fixes: 15ad9762d69f ("KVM: VMX: Save HOST_CR3 in vmx_prepare_switch_to_guest()")
Reported-by: Wanpeng Li <kernellwp@gmail.com>
Cc: Lai Jiangshan <laijs@linux.alibaba.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Acked-by: Lai Jiangshan <jiangshanlai@gmail.com>
Message-Id: <20220224191917.3508476-3-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoRevert "KVM: VMX: Save HOST_CR3 in vmx_set_host_fs_gs()"
Sean Christopherson [Thu, 24 Feb 2022 19:19:16 +0000 (19:19 +0000)]
Revert "KVM: VMX: Save HOST_CR3 in vmx_set_host_fs_gs()"

Undo a nested VMX fix as a step toward reverting the commit it fixed,
15ad9762d69f ("KVM: VMX: Save HOST_CR3 in vmx_prepare_switch_to_guest()"),
as the underlying premise that "host CR3 in the vcpu thread can only be
changed when scheduling" is wrong.

This reverts commit a9f2705ec84449e3b8d70c804766f8e97e23080d.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220224191917.3508476-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: nSVM: disallow userspace setting of MSR_AMD64_TSC_RATIO to non default...
Maxim Levitsky [Wed, 23 Feb 2022 11:56:49 +0000 (13:56 +0200)]
KVM: x86: nSVM: disallow userspace setting of MSR_AMD64_TSC_RATIO to non default value when tsc scaling disabled

If nested tsc scaling is disabled, MSR_AMD64_TSC_RATIO should
never have non default value.

Due to way nested tsc scaling support was implmented in qemu,
it would set this msr to 0 when nested tsc scaling was disabled.
Ignore that value for now, as it causes no harm.

Fixes: 5228eb96a487 ("KVM: x86: nSVM: implement nested TSC scaling")
Cc: stable@vger.kernel.org
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20220223115649.319134-1-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86/mmu: make apf token non-zero to fix bug
Liang Zhang [Tue, 22 Feb 2022 03:12:39 +0000 (11:12 +0800)]
KVM: x86/mmu: make apf token non-zero to fix bug

In current async pagefault logic, when a page is ready, KVM relies on
kvm_arch_can_dequeue_async_page_present() to determine whether to deliver
a READY event to the Guest. This function test token value of struct
kvm_vcpu_pv_apf_data, which must be reset to zero by Guest kernel when a
READY event is finished by Guest. If value is zero meaning that a READY
event is done, so the KVM can deliver another.
But the kvm_arch_setup_async_pf() may produce a valid token with zero
value, which is confused with previous mention and may lead the loss of
this READY event.

This bug may cause task blocked forever in Guest:
 INFO: task stress:7532 blocked for more than 1254 seconds.
       Not tainted 5.10.0 #16
 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
 task:stress          state:D stack:    0 pid: 7532 ppid:  1409
 flags:0x00000080
 Call Trace:
  __schedule+0x1e7/0x650
  schedule+0x46/0xb0
  kvm_async_pf_task_wait_schedule+0xad/0xe0
  ? exit_to_user_mode_prepare+0x60/0x70
  __kvm_handle_async_pf+0x4f/0xb0
  ? asm_exc_page_fault+0x8/0x30
  exc_page_fault+0x6f/0x110
  ? asm_exc_page_fault+0x8/0x30
  asm_exc_page_fault+0x1e/0x30
 RIP: 0033:0x402d00
 RSP: 002b:00007ffd31912500 EFLAGS: 00010206
 RAX: 0000000000071000 RBX: ffffffffffffffff RCX: 00000000021a32b0
 RDX: 000000000007d011 RSI: 000000000007d000 RDI: 00000000021262b0
 RBP: 00000000021262b0 R08: 0000000000000003 R09: 0000000000000086
 R10: 00000000000000eb R11: 00007fefbdf2baa0 R12: 0000000000000000
 R13: 0000000000000002 R14: 000000000007d000 R15: 0000000000001000

Signed-off-by: Liang Zhang <zhangliang5@huawei.com>
Message-Id: <20220222031239.1076682-1-zhangliang5@huawei.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoMerge branch 'kvm-ppc-cap-210' into kvm-master
Paolo Bonzini [Tue, 22 Feb 2022 14:07:16 +0000 (09:07 -0500)]
Merge branch 'kvm-ppc-cap-210' into kvm-master

By request of Nick Piggin:

> Patch 3 requires a KVM_CAP_PPC number allocated. QEMU maintainers are
> happy with it (link in changelog) just waiting on KVM upstreaming. Do
> you have objections to the series going to ppc/kvm tree first, or
> another option is you could take patch 3 alone first (it's relatively
> independent of the other 2) and ppc/kvm gets it from you?

2 years agoKVM: PPC: reserve capability 210 for KVM_CAP_PPC_AIL_MODE_3
Nicholas Piggin [Tue, 22 Feb 2022 14:06:54 +0000 (09:06 -0500)]
KVM: PPC: reserve capability 210 for KVM_CAP_PPC_AIL_MODE_3

Add KVM_CAP_PPC_AIL_MODE_3 to advertise the capability to set the AIL
resource mode to 3 with the H_SET_MODE hypercall. This capability
differs between processor types and KVM types (PR, HV, Nested HV), and
affects guest-visible behaviour.

QEMU will implement a cap-ail-mode-3 to control this behaviour[1], and
use the KVM CAP if available to determine KVM support[2].

Reviewed-by: Fabiano Rosas <farosas@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agox86/kvm: Don't use pv tlb/ipi/sched_yield if on 1 vCPU
Wanpeng Li [Fri, 18 Feb 2022 08:10:38 +0000 (00:10 -0800)]
x86/kvm: Don't use pv tlb/ipi/sched_yield if on 1 vCPU

Inspired by commit 3553ae5690a (x86/kvm: Don't use pvqspinlock code if
only 1 vCPU), on a VM with only 1 vCPU, there is no need to enable
pv tlb/ipi/sched_yield and we can save the memory for __pv_cpu_mask.

Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Message-Id: <1645171838-2855-1-git-send-email-wanpengli@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agox86/kvm: Fix compilation warning in non-x86_64 builds
Leonardo Bras [Fri, 18 Feb 2022 03:41:00 +0000 (00:41 -0300)]
x86/kvm: Fix compilation warning in non-x86_64 builds

On non-x86_64 builds, helpers gtod_is_based_on_tsc() and
kvm_guest_supported_xfd() are defined but never used.  Because these are
static inline but are in a .c file, some compilers do warn for them with
-Wunused-function, which becomes an error if -Werror is present.

Add #ifdef so they are only defined in x86_64 builds.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Message-Id: <20220218034100.115702-1-leobras@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agox86/kvm/fpu: Remove kvm_vcpu_arch.guest_supported_xcr0
Leonardo Bras [Thu, 17 Feb 2022 05:30:30 +0000 (02:30 -0300)]
x86/kvm/fpu: Remove kvm_vcpu_arch.guest_supported_xcr0

kvm_vcpu_arch currently contains the guest supported features in both
guest_supported_xcr0 and guest_fpu.fpstate->user_xfeatures field.

Currently both fields are set to the same value in
kvm_vcpu_after_set_cpuid() and are not changed anywhere else after that.

Since it's not good to keep duplicated data, remove guest_supported_xcr0.

To keep the code more readable, introduce kvm_guest_supported_xcr()
and kvm_guest_supported_xfd() to replace the previous usages of
guest_supported_xcr0.

Signed-off-by: Leonardo Bras <leobras@redhat.com>
Message-Id: <20220217053028.96432-3-leobras@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agox86/kvm/fpu: Limit guest user_xfeatures to supported bits of XCR0
Leonardo Bras [Thu, 17 Feb 2022 05:30:29 +0000 (02:30 -0300)]
x86/kvm/fpu: Limit guest user_xfeatures to supported bits of XCR0

During host/guest switch (like in kvm_arch_vcpu_ioctl_run()), the kernel
swaps the fpu between host/guest contexts, by using fpu_swap_kvm_fpstate().

When xsave feature is available, the fpu swap is done by:
- xsave(s) instruction, with guest's fpstate->xfeatures as mask, is used
  to store the current state of the fpu registers to a buffer.
- xrstor(s) instruction, with (fpu_kernel_cfg.max_features &
  XFEATURE_MASK_FPSTATE) as mask, is used to put the buffer into fpu regs.

For xsave(s) the mask is used to limit what parts of the fpu regs will
be copied to the buffer. Likewise on xrstor(s), the mask is used to
limit what parts of the fpu regs will be changed.

The mask for xsave(s), the guest's fpstate->xfeatures, is defined on
kvm_arch_vcpu_create(), which (in summary) sets it to all features
supported by the cpu which are enabled on kernel config.

This means that xsave(s) will save to guest buffer all the fpu regs
contents the cpu has enabled when the guest is paused, even if they
are not used.

This would not be an issue, if xrstor(s) would also do that.

xrstor(s)'s mask for host/guest swap is basically every valid feature
contained in kernel config, except XFEATURE_MASK_PKRU.
Accordingto kernel src, it is instead switched in switch_to() and
flush_thread().

Then, the following happens with a host supporting PKRU starts a
guest that does not support it:
1 - Host has XFEATURE_MASK_PKRU set. 1st switch to guest,
2 - xsave(s) fpu regs to host fpustate (buffer has XFEATURE_MASK_PKRU)
3 - xrstor(s) guest fpustate to fpu regs (fpu regs have XFEATURE_MASK_PKRU)
4 - guest runs, then switch back to host,
5 - xsave(s) fpu regs to guest fpstate (buffer now have XFEATURE_MASK_PKRU)
6 - xrstor(s) host fpstate to fpu regs.
7 - kvm_vcpu_ioctl_x86_get_xsave() copy guest fpstate to userspace (with
    XFEATURE_MASK_PKRU, which should not be supported by guest vcpu)

On 5, even though the guest does not support PKRU, it does have the flag
set on guest fpstate, which is transferred to userspace via vcpu ioctl
KVM_GET_XSAVE.

This becomes a problem when the user decides on migrating the above guest
to another machine that does not support PKRU: the new host restores
guest's fpu regs to as they were before (xrstor(s)), but since the new
host don't support PKRU, a general-protection exception ocurs in xrstor(s)
and that crashes the guest.

This can be solved by making the guest's fpstate->user_xfeatures hold
a copy of guest_supported_xcr0. This way, on 7 the only flags copied to
userspace will be the ones compatible to guest requirements, and thus
there will be no issue during migration.

As a bonus, it will also fail if userspace tries to set fpu features
(with the KVM_SET_XSAVE ioctl) that are not compatible to the guest
configuration.  Such features will never be returned by KVM_GET_XSAVE
or KVM_GET_XSAVE2.

Also, since kvm_vcpu_after_set_cpuid() now sets fpstate->user_xfeatures,
there is not need to set it in kvm_check_cpuid(). So, change
fpstate_realloc() so it does not touch fpstate->user_xfeatures if a
non-NULL guest_fpu is passed, which is the case when kvm_check_cpuid()
calls it.

Signed-off-by: Leonardo Bras <leobras@redhat.com>
Message-Id: <20220217053028.96432-2-leobras@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agokvm: x86: Disable KVM_HC_CLOCK_PAIRING if tsc is in always catchup mode
Anton Romanov [Wed, 16 Feb 2022 18:26:54 +0000 (18:26 +0000)]
kvm: x86: Disable KVM_HC_CLOCK_PAIRING if tsc is in always catchup mode

If vcpu has tsc_always_catchup set each request updates pvclock data.
KVM_HC_CLOCK_PAIRING consumers such as ptp_kvm_x86 rely on tsc read on
host's side and do hypercall inside pvclock_read_retry loop leading to
infinite loop in such situation.

v3:
    Removed warn
    Changed return code to KVM_EFAULT
v2:
    Added warn

Signed-off-by: Anton Romanov <romanton@google.com>
Message-Id: <20220216182653.506850-1-romanton@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: Fix lockdep false negative during host resume
Wanpeng Li [Tue, 15 Feb 2022 10:15:42 +0000 (02:15 -0800)]
KVM: Fix lockdep false negative during host resume

I saw the below splatting after the host suspended and resumed.

   WARNING: CPU: 0 PID: 2943 at kvm/arch/x86/kvm/../../../virt/kvm/kvm_main.c:5531 kvm_resume+0x2c/0x30 [kvm]
   CPU: 0 PID: 2943 Comm: step_after_susp Tainted: G        W IOE     5.17.0-rc3+ #4
   RIP: 0010:kvm_resume+0x2c/0x30 [kvm]
   Call Trace:
    <TASK>
    syscore_resume+0x90/0x340
    suspend_devices_and_enter+0xaee/0xe90
    pm_suspend.cold+0x36b/0x3c2
    state_store+0x82/0xf0
    kernfs_fop_write_iter+0x1b6/0x260
    new_sync_write+0x258/0x370
    vfs_write+0x33f/0x510
    ksys_write+0xc9/0x160
    do_syscall_64+0x3b/0xc0
    entry_SYSCALL_64_after_hwframe+0x44/0xae

lockdep_is_held() can return -1 when lockdep is disabled which triggers
this warning. Let's use lockdep_assert_not_held() which can detect
incorrect calls while holding a lock and it also avoids false negatives
when lockdep is disabled.

Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Message-Id: <1644920142-81249-1-git-send-email-wanpengli@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: Add KVM_CAP_ENABLE_CAP to x86
Aaron Lewis [Mon, 14 Feb 2022 21:29:51 +0000 (21:29 +0000)]
KVM: x86: Add KVM_CAP_ENABLE_CAP to x86

Follow the precedent set by other architectures that support the VCPU
ioctl, KVM_ENABLE_CAP, and advertise the VM extension, KVM_CAP_ENABLE_CAP.
This way, userspace can ensure that KVM_ENABLE_CAP is available on a
vcpu before using it.

Fixes: 5c919412fe61 ("kvm/x86: Hyper-V synthetic interrupt controller")
Signed-off-by: Aaron Lewis <aaronlewis@google.com>
Message-Id: <20220214212950.1776943-1-aaronlewis@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: arm64: Don't miss pending interrupts for suspended vCPU
Oliver Upton [Thu, 17 Feb 2022 10:12:42 +0000 (10:12 +0000)]
KVM: arm64: Don't miss pending interrupts for suspended vCPU

In order to properly emulate the WFI instruction, KVM reads back
ICH_VMCR_EL2 and enables doorbells for GICv4. These preparations are
necessary in order to recognize pending interrupts in
kvm_arch_vcpu_runnable() and return to the guest. Until recently, this
work was done by kvm_arch_vcpu_{blocking,unblocking}(). Since commit
6109c5a6ab7f ("KVM: arm64: Move vGIC v4 handling for WFI out arch
callback hook"), these callbacks were gutted and superseded by
kvm_vcpu_wfi().

It is important to note that KVM implements PSCI CPU_SUSPEND calls as
a WFI within the guest. However, the implementation calls directly into
kvm_vcpu_halt(), which skips the needed work done in kvm_vcpu_wfi()
to detect pending interrupts. Fix the issue by calling the WFI helper.

Fixes: 6109c5a6ab7f ("KVM: arm64: Move vGIC v4 handling for WFI out arch callback hook")
Signed-off-by: Oliver Upton <oupton@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220217101242.3013716-1-oupton@google.com
2 years agoKVM: x86/pmu: Use AMD64_RAW_EVENT_MASK for PERF_TYPE_RAW
Jim Mattson [Thu, 3 Feb 2022 01:48:13 +0000 (17:48 -0800)]
KVM: x86/pmu: Use AMD64_RAW_EVENT_MASK for PERF_TYPE_RAW

AMD's event select is 3 nybbles, with the high nybble in bits 35:32 of
a PerfEvtSeln MSR. Don't mask off the high nybble when configuring a
RAW perf event.

Fixes: ca724305a2b0 ("KVM: x86/vPMU: Implement AMD vPMU code for KVM")
Signed-off-by: Jim Mattson <jmattson@google.com>
Message-Id: <20220203014813.2130559-2-jmattson@google.com>
Reviewed-by: David Dunn <daviddunn@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86/pmu: Don't truncate the PerfEvtSeln MSR when creating a perf event
Jim Mattson [Thu, 3 Feb 2022 01:48:12 +0000 (17:48 -0800)]
KVM: x86/pmu: Don't truncate the PerfEvtSeln MSR when creating a perf event

AMD's event select is 3 nybbles, with the high nybble in bits 35:32 of
a PerfEvtSeln MSR. Don't drop the high nybble when setting up the
config field of a perf_event_attr structure for a call to
perf_event_create_kernel_counter().

Fixes: ca724305a2b0 ("KVM: x86/vPMU: Implement AMD vPMU code for KVM")
Reported-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Jim Mattson <jmattson@google.com>
Message-Id: <20220203014813.2130559-1-jmattson@google.com>
Reviewed-by: David Dunn <daviddunn@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: SVM: fix race between interrupt delivery and AVIC inhibition
Maxim Levitsky [Tue, 8 Feb 2022 11:48:42 +0000 (06:48 -0500)]
KVM: SVM: fix race between interrupt delivery and AVIC inhibition

If svm_deliver_avic_intr is called just after the target vcpu's AVIC got
inhibited, it might read a stale value of vcpu->arch.apicv_active
which can lead to the target vCPU not noticing the interrupt.

To fix this use load-acquire/store-release so that, if the target vCPU
is IN_GUEST_MODE, we're guaranteed to see a previous disabling of the
AVIC.  If AVIC has been disabled in the meanwhile, proceed with the
KVM_REQ_EVENT-based delivery.

Incomplete IPI vmexit has the same races as svm_deliver_avic_intr, and
in fact it can be handled in exactly the same way; the only difference
lies in who has set IRR, whether svm_deliver_interrupt or the processor.
Therefore, svm_complete_interrupt_delivery can be used to fix incomplete
IPI vmexits as well.

Co-developed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: SVM: set IRR in svm_deliver_interrupt
Paolo Bonzini [Tue, 8 Feb 2022 11:57:07 +0000 (06:57 -0500)]
KVM: SVM: set IRR in svm_deliver_interrupt

SVM has to set IRR for both the AVIC and the software-LAPIC case,
so pull it up to the common function that handles both configurations.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: SVM: extract avic_ring_doorbell
Maxim Levitsky [Tue, 8 Feb 2022 11:45:16 +0000 (06:45 -0500)]
KVM: SVM: extract avic_ring_doorbell

The check on the current CPU adds an extra level of indentation to
svm_deliver_avic_intr and conflates documentation on what happens
if the vCPU exits (of interest to svm_deliver_avic_intr) and migrates
(only of interest to avic_ring_doorbell, which calls get/put_cpu()).
Extract the wrmsr to a separate function and rewrite the
comment in svm_deliver_avic_intr().

Co-developed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoselftests: kvm: Remove absent target file
Muhammad Usama Anjum [Thu, 10 Feb 2022 17:23:51 +0000 (22:23 +0500)]
selftests: kvm: Remove absent target file

There is no vmx_pi_mmio_test file. Remove it to get rid of error while
creation of selftest archive:

rsync: [sender] link_stat "/kselftest/kvm/x86_64/vmx_pi_mmio_test" failed: No such file or directory (2)
rsync error: some files/attrs were not transferred (see previous errors) (code 23) at main.c(1333) [sender=3.2.3]

Fixes: 6a58150859fd ("selftest: KVM: Add intra host migration tests")
Reported-by: "kernelci.org bot" <bot@kernelci.org>
Signed-off-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Message-Id: <20220210172352.1317554-1-usama.anjum@collabora.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoMerge tag 'kvmarm-fixes-5.17-3' of git://git.kernel.org/pub/scm/linux/kernel/git...
Paolo Bonzini [Fri, 11 Feb 2022 17:10:57 +0000 (12:10 -0500)]
Merge tag 'kvmarm-fixes-5.17-3' of git://git./linux/kernel/git/kvmarm/kvmarm into HEAD

KVM/arm64 fixes for 5.17, take #3

- Fix pending state read of a HW interrupt

2 years agoKVM: arm64: vgic: Read HW interrupt pending state from the HW
Marc Zyngier [Thu, 3 Feb 2022 09:24:45 +0000 (09:24 +0000)]
KVM: arm64: vgic: Read HW interrupt pending state from the HW

It appears that a read access to GIC[DR]_I[CS]PENDRn doesn't always
result in the pending interrupts being accurately reported if they are
mapped to a HW interrupt. This is particularily visible when acking
the timer interrupt and reading the GICR_ISPENDR1 register immediately
after, for example (the interrupt appears as not-pending while it really
is...).

This is because a HW interrupt has its 'active and pending state' kept
in the *physical* distributor, and not in the virtual one, as mandated
by the spec (this is what allows the direct deactivation). The virtual
distributor only caries the pending and active *states* (note the
plural, as these are two independent and non-overlapping states).

Fix it by reading the HW state back, either from the timer itself or
from the distributor if necessary.

Reported-by: Ricardo Koller <ricarkol@google.com>
Tested-by: Ricardo Koller <ricarkol@google.com>
Reviewed-by: Ricardo Koller <ricarkol@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220208123726.3604198-1-maz@kernel.org
2 years agoKVM: x86/xen: Fix runstate updates to be atomic when preempting vCPU
David Woodhouse [Mon, 25 Oct 2021 13:29:01 +0000 (14:29 +0100)]
KVM: x86/xen: Fix runstate updates to be atomic when preempting vCPU

There are circumstances whem kvm_xen_update_runstate_guest() should not
sleep because it ends up being called from __schedule() when the vCPU
is preempted:

[  222.830825]  kvm_xen_update_runstate_guest+0x24/0x100
[  222.830878]  kvm_arch_vcpu_put+0x14c/0x200
[  222.830920]  kvm_sched_out+0x30/0x40
[  222.830960]  __schedule+0x55c/0x9f0

To handle this, make it use the same trick as __kvm_xen_has_interrupt(),
of using the hva from the gfn_to_hva_cache directly. Then it can use
pagefault_disable() around the accesses and just bail out if the page
is absent (which is unlikely).

I almost switched to using a gfn_to_pfn_cache here and bailing out if
kvm_map_gfn() fails, like kvm_steal_time_set_preempted() does â€” but on
closer inspection it looks like kvm_map_gfn() will *always* fail in
atomic context for a page in IOMEM, which means it will silently fail
to make the update every single time for such guests, AFAICT. So I
didn't do it that way after all. And will probably fix that one too.

Cc: stable@vger.kernel.org
Fixes: 30b5c851af79 ("KVM: x86/xen: Add support for vCPU runstate information")
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Message-Id: <b17a93e5ff4561e57b1238e3e7ccd0b613eb827e.camel@infradead.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: SVM: move avic definitions from AMD's spec to svm.h
Maxim Levitsky [Mon, 7 Feb 2022 15:54:26 +0000 (17:54 +0200)]
KVM: x86: SVM: move avic definitions from AMD's spec to svm.h

asm/svm.h is the correct place for all values that are defined in
the SVM spec, and that includes AVIC.

Also add some values from the spec that were not defined before
and will be soon useful.

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20220207155447.840194-10-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: lapic: don't touch irr_pending in kvm_apic_update_apicv when inhibiting it
Maxim Levitsky [Mon, 7 Feb 2022 15:54:25 +0000 (17:54 +0200)]
KVM: x86: lapic: don't touch irr_pending in kvm_apic_update_apicv when inhibiting it

kvm_apic_update_apicv is called when AVIC is still active, thus IRR bits
can be set by the CPU after it is called, and don't cause the irr_pending
to be set to true.

Also logic in avic_kick_target_vcpu doesn't expect a race with this
function so to make it simple, just keep irr_pending set to true and
let the next interrupt injection to the guest clear it.

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20220207155447.840194-9-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: nSVM: deal with L1 hypervisor that intercepts interrupts but lets L2 contro...
Maxim Levitsky [Mon, 7 Feb 2022 15:54:24 +0000 (17:54 +0200)]
KVM: x86: nSVM: deal with L1 hypervisor that intercepts interrupts but lets L2 control them

Fix a corner case in which the L1 hypervisor intercepts
interrupts (INTERCEPT_INTR) and either doesn't set
virtual interrupt masking (V_INTR_MASKING) or enters a
nested guest with EFLAGS.IF disabled prior to the entry.

In this case, despite the fact that L1 intercepts the interrupts,
KVM still needs to set up an interrupt window to wait before
injecting the INTR vmexit.

Currently the KVM instead enters an endless loop of 'req_immediate_exit'.

Exactly the same issue also happens for SMIs and NMI.
Fix this as well.

Note that on VMX, this case is impossible as there is only
'vmexit on external interrupts' execution control which either set,
in which case both host and guest's EFLAGS.IF
are ignored, or not set, in which case no VMexits are delivered.

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20220207155447.840194-8-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: nSVM: expose clean bit support to the guest
Maxim Levitsky [Mon, 7 Feb 2022 15:54:22 +0000 (17:54 +0200)]
KVM: x86: nSVM: expose clean bit support to the guest

KVM already honours few clean bits thus it makes sense
to let the nested guest know about it.

Note that KVM also doesn't check if the hardware supports
clean bits, and therefore nested KVM was
already setting clean bits and L0 KVM
was already honouring them.

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20220207155447.840194-6-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: nSVM/nVMX: set nested_run_pending on VM entry which is a result of RSM
Maxim Levitsky [Mon, 7 Feb 2022 15:54:21 +0000 (17:54 +0200)]
KVM: x86: nSVM/nVMX: set nested_run_pending on VM entry which is a result of RSM

While RSM induced VM entries are not full VM entries,
they still need to be followed by actual VM entry to complete it,
unlike setting the nested state.

This patch fixes boot of hyperv and SMM enabled
windows VM running nested on KVM, which fail due
to this issue combined with lack of dirty bit setting.

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Cc: stable@vger.kernel.org
Message-Id: <20220207155447.840194-5-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: nSVM: mark vmcb01 as dirty when restoring SMM saved state
Maxim Levitsky [Mon, 7 Feb 2022 15:54:20 +0000 (17:54 +0200)]
KVM: x86: nSVM: mark vmcb01 as dirty when restoring SMM saved state

While usually, restoring the smm state makes the KVM enter
the nested guest thus a different vmcb (vmcb02 vs vmcb01),
KVM should still mark it as dirty, since hardware
can in theory cache multiple vmcbs.

Failure to do so, combined with lack of setting the
nested_run_pending (which is fixed in the next patch),
might make KVM re-enter vmcb01, which was just exited from,
with completely different set of guest state registers
(SMM vs non SMM) and without proper dirty bits set,
which results in the CPU reusing stale IDTR pointer
which leads to a guest shutdown on any interrupt.

On the real hardware this usually doesn't happen,
but when running nested, L0's KVM does check and
honour few dirty bits, causing this issue to happen.

This patch fixes boot of hyperv and SMM enabled
windows VM running nested on KVM.

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Cc: stable@vger.kernel.org
Message-Id: <20220207155447.840194-4-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: nSVM: fix potential NULL derefernce on nested migration
Maxim Levitsky [Mon, 7 Feb 2022 15:54:19 +0000 (17:54 +0200)]
KVM: x86: nSVM: fix potential NULL derefernce on nested migration

Turns out that due to review feedback and/or rebases
I accidentally moved the call to nested_svm_load_cr3 to be too early,
before the NPT is enabled, which is very wrong to do.

KVM can't even access guest memory at that point as nested NPT
is needed for that, and of course it won't initialize the walk_mmu,
which is main issue the patch was addressing.

Fix this for real.

Fixes: 232f75d3b4b5 ("KVM: nSVM: call nested_svm_load_cr3 on nested state load")
Cc: stable@vger.kernel.org
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20220207155447.840194-3-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: SVM: don't passthrough SMAP/SMEP/PKE bits in !NPT && !gCR0.PG case
Maxim Levitsky [Mon, 7 Feb 2022 15:54:18 +0000 (17:54 +0200)]
KVM: x86: SVM: don't passthrough SMAP/SMEP/PKE bits in !NPT && !gCR0.PG case

When the guest doesn't enable paging, and NPT/EPT is disabled, we
use guest't paging CR3's as KVM's shadow paging pointer and
we are technically in direct mode as if we were to use NPT/EPT.

In direct mode we create SPTEs with user mode permissions
because usually in the direct mode the NPT/EPT doesn't
need to restrict access based on guest CPL
(there are MBE/GMET extenstions for that but KVM doesn't use them).

In this special "use guest paging as direct" mode however,
and if CR4.SMAP/CR4.SMEP are enabled, that will make the CPU
fault on each access and KVM will enter endless loop of page faults.

Since page protection doesn't have any meaning in !PG case,
just don't passthrough these bits.

The fix is the same as was done for VMX in commit:
commit 656ec4a4928a ("KVM: VMX: fix SMEP and SMAP without EPT")

This fixes the boot of windows 10 without NPT for good.
(Without this patch, BSP boots, but APs were stuck in endless
loop of page faults, causing the VM boot with 1 CPU)

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Cc: stable@vger.kernel.org
Message-Id: <20220207155447.840194-2-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoRevert "svm: Add warning message for AVIC IPI invalid target"
Sean Christopherson [Fri, 4 Feb 2022 21:41:55 +0000 (21:41 +0000)]
Revert "svm: Add warning message for AVIC IPI invalid target"

Remove a WARN on an "AVIC IPI invalid target" exit, the WARN is trivial
to trigger from guest as it will fail on any destination APIC ID that
doesn't exist from the guest's perspective.

Don't bother recording anything in the kernel log, the common tracepoint
for kvm_avic_incomplete_ipi() is sufficient for debugging.

This reverts commit 37ef0c4414c9743ba7f1af4392f0a27a99649f2a.

Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220204214205.3306634-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoLinux 5.17-rc3
Linus Torvalds [Sun, 6 Feb 2022 20:20:50 +0000 (12:20 -0800)]
Linux 5.17-rc3

2 years agoMerge tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 6 Feb 2022 18:34:45 +0000 (10:34 -0800)]
Merge tag 'ext4_for_linus_stable' of git://git./linux/kernel/git/tytso/ext4

Pull ext4 fixes from Ted Ts'o:
 "Various bug fixes for ext4 fast commit and inline data handling.

  Also fix regression introduced as part of moving to the new mount API"

* tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  fs/ext4: fix comments mentioning i_mutex
  ext4: fix incorrect type issue during replay_del_range
  jbd2: fix kernel-doc descriptions for jbd2_journal_shrink_{scan,count}()
  ext4: fix potential NULL pointer dereference in ext4_fill_super()
  jbd2: refactor wait logic for transaction updates into a common function
  jbd2: cleanup unused functions declarations from jbd2.h
  ext4: fix error handling in ext4_fc_record_modified_inode()
  ext4: remove redundant max inline_size check in ext4_da_write_inline_data_begin()
  ext4: fix error handling in ext4_restore_inline_data()
  ext4: fast commit may miss file actions
  ext4: fast commit may not fallback for ineligible commit
  ext4: modify the logic of ext4_mb_new_blocks_simple
  ext4: prevent used blocks from being allocated during fast commit replay

2 years agoMerge tag 'perf-tools-fixes-for-v5.17-2022-02-06' of git://git.kernel.org/pub/scm...
Linus Torvalds [Sun, 6 Feb 2022 18:18:23 +0000 (10:18 -0800)]
Merge tag 'perf-tools-fixes-for-v5.17-2022-02-06' of git://git./linux/kernel/git/acme/linux

Pull perf tools fixes from Arnaldo Carvalho de Melo:

 - Fix display of grouped aliased events in 'perf stat'.

 - Add missing branch_sample_type to perf_event_attr__fprintf().

 - Apply correct label to user/kernel symbols in branch mode.

 - Fix 'perf ftrace' system_wide tracing, it has to be set before
   creating the maps.

 - Return error if procfs isn't mounted for PID namespaces when
   synthesizing records for pre-existing processes.

 - Set error stream of objdump process for 'perf annotate' TUI, to avoid
   garbling the screen.

 - Add missing arm64 support to perf_mmap__read_self(), the kernel part
   got into 5.17.

 - Check for NULL pointer before dereference writing debug info about a
   sample.

 - Update UAPI copies for asound, perf_event, prctl and kvm headers.

 - Fix a typo in bpf_counter_cgroup.c.

* tag 'perf-tools-fixes-for-v5.17-2022-02-06' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
  perf ftrace: system_wide collection is not effective by default
  libperf: Add arm64 support to perf_mmap__read_self()
  tools include UAPI: Sync sound/asound.h copy with the kernel sources
  perf stat: Fix display of grouped aliased events
  perf tools: Apply correct label to user/kernel symbols in branch mode
  perf bpf: Fix a typo in bpf_counter_cgroup.c
  perf synthetic-events: Return error if procfs isn't mounted for PID namespaces
  perf session: Check for NULL pointer before dereference
  perf annotate: Set error stream of objdump process for TUI
  perf tools: Add missing branch_sample_type to perf_event_attr__fprintf()
  tools headers UAPI: Sync linux/kvm.h with the kernel sources
  tools headers UAPI: Sync linux/prctl.h with the kernel sources
  perf beauty: Make the prctl arg regexp more strict to cope with PR_SET_VMA
  tools headers cpufeatures: Sync with the kernel sources
  tools headers UAPI: Sync linux/perf_event.h with the kernel sources
  tools include UAPI: Sync sound/asound.h copy with the kernel sources

2 years agoMerge tag 'perf_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 6 Feb 2022 18:11:14 +0000 (10:11 -0800)]
Merge tag 'perf_urgent_for_v5.17_rc3' of git://git./linux/kernel/git/tip/tip

Pull perf fixes from Borislav Petkov:

 - Intel/PT: filters could crash the kernel

 - Intel: default disable the PMU for SMM, some new-ish EFI firmware has
   started using CPL3 and the PMU CPL filters don't discriminate against
   SMM, meaning that CPL3 (userspace only) events now also count EFI/SMM
   cycles.

 - Fixup for perf_event_attr::sig_data

* tag 'perf_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/x86/intel/pt: Fix crash with stop filters in single-range mode
  perf: uapi: Document perf_event_attr::sig_data truncation on 32 bit architectures
  selftests/perf_events: Test modification of perf_event_attr::sig_data
  perf: Copy perf_event_attr::sig_data on modification
  x86/perf: Default set FREEZE_ON_SMI for all

2 years agoMerge tag 'objtool_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 6 Feb 2022 18:04:43 +0000 (10:04 -0800)]
Merge tag 'objtool_urgent_for_v5.17_rc3' of git://git./linux/kernel/git/tip/tip

Pull objtool fix from Borislav Petkov:
 "Fix a potential truncated string warning triggered by gcc12"

* tag 'objtool_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  objtool: Fix truncated string warning

2 years agoMerge tag 'irq_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 6 Feb 2022 18:00:40 +0000 (10:00 -0800)]
Merge tag 'irq_urgent_for_v5.17_rc3' of git://git./linux/kernel/git/tip/tip

Pull irq fix from Borislav Petkov:
 "Remove a bogus warning introduced by the recent PCI MSI irq affinity
  overhaul"

* tag 'irq_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  PCI/MSI: Remove bogus warning in pci_irq_get_affinity()

2 years agoMerge tag 'edac_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 6 Feb 2022 17:57:39 +0000 (09:57 -0800)]
Merge tag 'edac_urgent_for_v5.17_rc3' of git://git./linux/kernel/git/ras/ras

Pull EDAC fixes from Borislav Petkov:
 "Fix altera and xgene EDAC drivers to propagate the correct error code
  from platform_get_irq() so that deferred probing still works"

* tag 'edac_urgent_for_v5.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
  EDAC/xgene: Fix deferred probing
  EDAC/altera: Fix deferred probing

2 years agoperf ftrace: system_wide collection is not effective by default
Changbin Du [Thu, 27 Jan 2022 13:20:10 +0000 (21:20 +0800)]
perf ftrace: system_wide collection is not effective by default

The ftrace.target.system_wide must be set before invoking
evlist__create_maps(), otherwise it has no effect.

Fixes: 53be50282269b46c ("perf ftrace: Add 'latency' subcommand")
Signed-off-by: Changbin Du <changbin.du@gmail.com>
Acked-by: Namhyung Kim <namhyung@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220127132010.4836-1-changbin.du@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agolibperf: Add arm64 support to perf_mmap__read_self()
Rob Herring [Tue, 1 Feb 2022 21:40:56 +0000 (15:40 -0600)]
libperf: Add arm64 support to perf_mmap__read_self()

Add the arm64 variants for read_perf_counter() and read_timestamp().
Unfortunately the counter number is encoded into the instruction, so the
code is a bit verbose to enumerate all possible counters.

Tested-by: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Tested-by: John Garry <john.garry@huawei.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Link: https://lore.kernel.org/r/20220201214056.702854-1-robh@kernel.org
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Will Deacon <will@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-perf-users@vger.kernel.org
2 years agotools include UAPI: Sync sound/asound.h copy with the kernel sources
Arnaldo Carvalho de Melo [Wed, 12 Feb 2020 14:04:23 +0000 (11:04 -0300)]
tools include UAPI: Sync sound/asound.h copy with the kernel sources

Picking the changes from:

  06feec6005c9d950 ("ASoC: hdmi-codec: Fix OOB memory accesses")

Which entails no changes in the tooling side as it doesn't introduce new
SNDRV_PCM_IOCTL_ ioctls.

To silence this perf tools build warning:

  Warning: Kernel ABI header at 'tools/include/uapi/sound/asound.h' differs from latest version at 'include/uapi/sound/asound.h'
  diff -u tools/include/uapi/sound/asound.h include/uapi/sound/asound.h

Cc: Dmitry Osipenko <digetx@gmail.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/lkml/Yf+6OT+2eMrYDEeX@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoperf stat: Fix display of grouped aliased events
Ian Rogers [Sat, 5 Feb 2022 01:09:41 +0000 (17:09 -0800)]
perf stat: Fix display of grouped aliased events

An event may have a number of uncore aliases that when added to the
evlist are consecutive.

If there are multiple uncore events in a group then
parse_events__set_leader_for_uncore_aliase will reorder the evlist so
that events on the same PMU are adjacent.

The collect_all_aliases function assumes that aliases are in blocks so
that only the first counter is printed and all others are marked merged.

The reordering for groups breaks the assumption and so all counts are
printed.

This change removes the assumption from collect_all_aliases
that the events are in blocks and instead processes the entire evlist.

Before:

  ```
  $ perf stat -e '{UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE,UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE},duration_time' -a -A -- sleep 1

   Performance counter stats for 'system wide':

  CPU0                  256,866      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 494,413      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                      967      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,738      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  285,161      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 429,920      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                      955      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,443      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  310,753      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 416,657      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,231      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,573      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  416,067      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 405,966      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,481      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,447      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  312,911      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 408,154      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,086      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,380      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  333,994      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 370,349      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,287      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,335      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  188,107      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 302,423      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                      701      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,070      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  307,221      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 383,642      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,036      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,158      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  318,479      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 821,545      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,028      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   2,550      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  227,618      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 372,272      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                      903      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,456      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  376,783      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 419,827      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,406      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,453      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  286,583      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 429,956      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                      999      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,436      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  313,867      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 370,159      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,114      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,291      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  342,083      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 409,111      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,399      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,684      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  365,828      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 376,037      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,378      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,411      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  382,456      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 621,743      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,232      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,955      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  342,316      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 385,067      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,176      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,268      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  373,588      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 386,163      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,394      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,464      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  381,206      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 546,891      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,266      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,712      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  221,176      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 392,069      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                      831      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,456      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  355,401      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 705,595      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,235      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   2,216      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  371,436      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 428,103      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,306      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,442      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  384,352      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 504,200      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,468      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,860      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  228,856      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 287,976      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                      832      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,060      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  215,121      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 334,162      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                      681      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,026      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  296,179      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 436,083      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,084      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,525      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  262,296      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 416,573      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                      986      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,533      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  285,852      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 359,842      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,073      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,326      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  303,379      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 367,222      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,008      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,156      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  273,487      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 425,449      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                      932      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,367      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  297,596      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 414,793      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,140      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,601      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  342,365      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 360,422      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,291      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,342      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  327,196      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 580,858      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,122      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   2,014      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  296,564      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 452,817      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,087      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,694      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  375,002      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 389,393      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,478      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   1,540      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0                  365,213      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36                 594,685      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                    1,401      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                   2,222      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0            1,000,749,060 ns   duration_time

         1.000749060 seconds time elapsed
  ```

After:

  ```
   Performance counter stats for 'system wide':

  CPU0               20,547,434      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU36              45,202,862      UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE
  CPU0                   82,001      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU36                 159,688      UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE
  CPU0            1,000,464,828 ns   duration_time

         1.000464828 seconds time elapsed
  ```

Fixes: 3cdc5c2cb924acb4 ("perf parse-events: Handle uncore event aliases in small groups properly")
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexandre Torgue <alexandre.torgue@foss.st.com>
Cc: Asaf Yaffe <asaf.yaffe@intel.com>
Cc: Caleb Biggers <caleb.biggers@intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Kshipra Bopardikar <kshipra.bopardikar@intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Perry Taylor <perry.taylor@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Vineet Singh <vineet.singh@intel.com>
Cc: Zhengjun Xing <zhengjun.xing@linux.intel.com>
Link: https://lore.kernel.org/r/20220205010941.1065469-1-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoperf tools: Apply correct label to user/kernel symbols in branch mode
German Gomez [Wed, 26 Jan 2022 10:59:26 +0000 (10:59 +0000)]
perf tools: Apply correct label to user/kernel symbols in branch mode

In branch mode, the branch symbols were being displayed with incorrect
cpumode labels. So fix this.

For example, before:
  # perf record -b -a -- sleep 1
  # perf report -b

  Overhead  Command  Source Shared Object  Source Symbol               Target Symbol
     0.08%  swapper  [kernel.kallsyms]     [k] rcu_idle_enter          [k] cpuidle_enter_state
 ==> 0.08%  cmd0     [kernel.kallsyms]     [.] psi_group_change        [.] psi_group_change
     0.08%  cmd1     [kernel.kallsyms]     [k] psi_group_change        [k] psi_group_change

After:
  # perf report -b

  Overhead  Command  Source Shared Object  Source Symbol               Target Symbol
     0.08%  swapper  [kernel.kallsyms]     [k] rcu_idle_enter          [k] cpuidle_enter_state
     0.08%  cmd0     [kernel.kallsyms]     [k] psi_group_change        [k] pei_group_change
     0.08%  cmd1     [kernel.kallsyms]     [k] psi_group_change        [k] psi_group_change

Reviewed-by: James Clark <james.clark@arm.com>
Signed-off-by: German Gomez <german.gomez@arm.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20220126105927.3411216-1-german.gomez@arm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoperf bpf: Fix a typo in bpf_counter_cgroup.c
Masanari Iida [Sat, 25 Dec 2021 00:55:58 +0000 (09:55 +0900)]
perf bpf: Fix a typo in bpf_counter_cgroup.c

This patch fixes a spelling typo in error message.

Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211225005558.503935-1-standby24x7@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoperf synthetic-events: Return error if procfs isn't mounted for PID namespaces
Leo Yan [Fri, 24 Dec 2021 12:40:13 +0000 (20:40 +0800)]
perf synthetic-events: Return error if procfs isn't mounted for PID namespaces

For perf recording, it retrieves process info by iterating nodes in proc
fs.  If we run perf in a non-root PID namespace with command:

  # unshare --fork --pid perf record -e cycles -a -- test_program

... in this case, unshare command creates a child PID namespace and
launches perf tool in it, but the issue is the proc fs is not mounted
for the non-root PID namespace, this leads to the perf tool gathering
process info from its parent PID namespace.

We can use below command to observe the process nodes under proc fs:

  # unshare --pid --fork ls /proc
1    137   1968  2128  3    342  48  62   78      crypto   kcore        net       uptime
10   138   2  2142  30   35  49  63   8      devices   keys        pagetypeinfo   version
11   139   20  2143  304  36  50  64   82      device-tree  key-users    partitions     vmallocinfo
12   14    2011  22    305  37  51  65   83      diskstats   kmsg        self       vmstat
128  140   2038  23    307  39  52  656  84      driver   kpagecgroup  slabinfo       zoneinfo
129  15    2074  24    309  4  53  67   9      execdomains  kpagecount   softirqs
13   16    2094  241   31   40  54  68   asound     fb   kpageflags   stat
130  164   2096  242   310  41  55  69   buddyinfo  filesystems  loadavg      swaps
131  17    2098  25    317  42  56  70   bus      fs   locks        sys
132  175   21  26    32   43  57  71   cgroups    interrupts   meminfo      sysrq-trigger
133  179   2102  263   329  44  58  75   cmdline    iomem   misc        sysvipc
134  1875  2103  27    330  45  59  76   config.gz  ioports   modules      thread-self
135  19    2117  29    333  46  6   77   consoles   irq   mounts       timer_list
136  1941  2121  298   34   47  60  773  cpuinfo    kallsyms   mtd        tty

So it shows many existed tasks, since unshared command has not mounted
the proc fs for the new created PID namespace, it still accesses the
proc fs of the root PID namespace.  This leads to two prominent issues:

- Firstly, PID values are mismatched between thread info and samples.
  The gathered thread info are coming from the proc fs of the root PID
  namespace, but samples record its PID from the child PID namespace.

- The second issue is profiled program 'test_program' returns its forked
  PID number from the child PID namespace, perf tool wrongly uses this
  PID number to retrieve the process info via the proc fs of the root
  PID namespace.

To avoid issues, we need to mount proc fs for the child PID namespace
with the option '--mount-proc' when use unshare command:

  # unshare --fork --pid --mount-proc perf record -e cycles -a -- test_program

Conversely, when the proc fs of the root PID namespace is used by child
namespace, perf tool can detect the multiple PID levels and
nsinfo__is_in_root_namespace() returns false, this patch reports error
for this case:

  # unshare --fork --pid perf record -e cycles -a -- test_program
  Couldn't synthesize bpf events.
  Perf runs in non-root PID namespace but it tries to gather process info from its parent PID namespace.
  Please mount the proc file system properly, e.g. add the option '--mount-proc' for unshare command.

Reviewed-by: James Clark <james.clark@arm.com>
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: KP Singh <kpsingh@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Song Liu <songliubraving@fb.com>
Cc: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/20211224124014.2492751-1-leo.yan@linaro.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoperf session: Check for NULL pointer before dereference
Ameer Hamza [Tue, 25 Jan 2022 12:11:41 +0000 (17:11 +0500)]
perf session: Check for NULL pointer before dereference

Move NULL pointer check before dereferencing the variable.

Addresses-Coverity: 1497622 ("Derereference before null check")
Reviewed-by: James Clark <james.clark@arm.com>
Signed-off-by: Ameer Hamza <amhamza.mgc@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
Cc: German Gomez <german.gomez@arm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Leo Yan <leo.yan@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Riccardo Mancini <rickyman7@gmail.com>
Link: https://lore.kernel.org/r/20220125121141.18347-1-amhamza.mgc@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoperf annotate: Set error stream of objdump process for TUI
Namhyung Kim [Wed, 2 Feb 2022 07:08:25 +0000 (23:08 -0800)]
perf annotate: Set error stream of objdump process for TUI

The stderr should be set to a pipe when using TUI.  Otherwise it'd
print to stdout and break TUI windows with an error message.

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20220202070828.143303-2-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoperf tools: Add missing branch_sample_type to perf_event_attr__fprintf()
Anshuman Khandual [Wed, 2 Feb 2022 10:57:23 +0000 (16:27 +0530)]
perf tools: Add missing branch_sample_type to perf_event_attr__fprintf()

This updates branch sample type with missing PERF_SAMPLE_BRANCH_TYPE_SAVE.

Suggested-by: James Clark <james.clark@arm.com>
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lore.kernel.org/lkml/1643799443-15109-1-git-send-email-anshuman.khandual@arm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agotools headers UAPI: Sync linux/kvm.h with the kernel sources
Arnaldo Carvalho de Melo [Sun, 9 May 2021 12:39:02 +0000 (09:39 -0300)]
tools headers UAPI: Sync linux/kvm.h with the kernel sources

To pick the changes in:

  f6c6804c43fa18d3 ("kvm: Move KVM_GET_XSAVE2 IOCTL definition at the end of kvm.h")

That just rebuilds perf, as these patches don't add any new KVM ioctl to
be harvested for the the 'perf trace' ioctl syscall argument
beautifiers.

This is also by now used by tools/testing/selftests/kvm/, a simple test
build succeeded.

This silences this perf build warning:

  Warning: Kernel ABI header at 'tools/include/uapi/linux/kvm.h' differs from latest version at 'include/uapi/linux/kvm.h'
  diff -u tools/include/uapi/linux/kvm.h include/uapi/linux/kvm.h

Cc: Janosch Frank <frankja@linux.ibm.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Link: http://lore.kernel.org/lkml/Yf+4k5Fs5Q3HdSG9@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoMerge remote-tracking branch 'torvalds/master' into perf/urgent
Arnaldo Carvalho de Melo [Sun, 6 Feb 2022 11:28:34 +0000 (08:28 -0300)]
Merge remote-tracking branch 'torvalds/master' into perf/urgent

To check if more kernel API sync is needed and also to see if the perf
build tests continue to pass.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2 years agoMerge tag 'for-linus-5.17a-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 5 Feb 2022 18:40:17 +0000 (10:40 -0800)]
Merge tag 'for-linus-5.17a-rc3-tag' of git://git./linux/kernel/git/xen/tip

Pull xen fixes from Juergen Gross:

 - documentation fixes related to Xen

 - enable x2apic mode when available when running as hardware
   virtualized guest under Xen

 - cleanup and fix a corner case of vcpu enumeration when running a
   paravirtualized Xen guest

* tag 'for-linus-5.17a-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  x86/Xen: streamline (and fix) PV CPU enumeration
  xen: update missing ioctl magic numers documentation
  Improve docs for IOCTL_GNTDEV_MAP_GRANT_REF
  xen: xenbus_dev.h: delete incorrect file name
  xen/x2apic: enable x2apic mode when supported for HVM

2 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Linus Torvalds [Sat, 5 Feb 2022 17:55:59 +0000 (09:55 -0800)]
Merge tag 'for-linus' of git://git./virt/kvm/kvm

Pull kvm fixes from Paolo Bonzini:
 "ARM:

   - A couple of fixes when handling an exception while a SError has
     been delivered

   - Workaround for Cortex-A510's single-step erratum

  RISC-V:

   - Make CY, TM, and IR counters accessible in VU mode

   - Fix SBI implementation version

  x86:

   - Report deprecation of x87 features in supported CPUID

   - Preparation for fixing an interrupt delivery race on AMD hardware

   - Sparse fix

  All except POWER and s390:

   - Rework guest entry code to correctly mark noinstr areas and fix
     vtime' accounting (for x86, this was already mostly correct but not
     entirely; for ARM, MIPS and RISC-V it wasn't)"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: x86: Use ERR_PTR_USR() to return -EFAULT as a __user pointer
  KVM: x86: Report deprecated x87 features in supported CPUID
  KVM: arm64: Workaround Cortex-A510's single-step and PAC trap errata
  KVM: arm64: Stop handle_exit() from handling HVC twice when an SError occurs
  KVM: arm64: Avoid consuming a stale esr value when SError occur
  RISC-V: KVM: Fix SBI implementation version
  RISC-V: KVM: make CY, TM, and IR counters accessible in VU mode
  kvm/riscv: rework guest entry logic
  kvm/arm64: rework guest entry logic
  kvm/x86: rework guest entry logic
  kvm/mips: rework guest entry logic
  kvm: add guest_state_{enter,exit}_irqoff()
  KVM: x86: Move delivery of non-APICv interrupt into vendor code
  kvm: Move KVM_GET_XSAVE2 IOCTL definition at the end of kvm.h

2 years agoMerge tag 'xfs-5.17-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Linus Torvalds [Sat, 5 Feb 2022 17:21:55 +0000 (09:21 -0800)]
Merge tag 'xfs-5.17-fixes-1' of git://git./fs/xfs/xfs-linux

Pull xfs fixes from Darrick Wong:
 "I was auditing operations in XFS that clear file privileges, and
  realized that XFS' fallocate implementation drops suid/sgid but
  doesn't clear file capabilities the same way that file writes and
  reflink do.

  There are VFS helpers that do it correctly, so refactor XFS to use
  them. I also noticed that we weren't flushing the log at the correct
  point in the fallocate operation, so that's fixed too.

  Summary:

   - Fix fallocate so that it drops all file privileges when files are
     modified instead of open-coding that incompletely.

   - Fix fallocate to flush the log if the caller wanted synchronous
     file updates"

* tag 'xfs-5.17-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: ensure log flush at the end of a synchronous fallocate call
  xfs: move xfs_update_prealloc_flags() to xfs_pnfs.c
  xfs: set prealloc flag in xfs_alloc_file_space()
  xfs: fallocate() should call file_modified()
  xfs: remove XFS_PREALLOC_SYNC
  xfs: reject crazy array sizes being fed to XFS_IOC_GETBMAP*

2 years agoMerge tag 'vfs-5.17-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Linus Torvalds [Sat, 5 Feb 2022 17:13:51 +0000 (09:13 -0800)]
Merge tag 'vfs-5.17-fixes-2' of git://git./fs/xfs/xfs-linux

Pull vfs fixes from Darrick Wong:
 "I was auditing the sync_fs code paths recently and noticed that most
  callers of ->sync_fs ignore its return value (and many implementations
  never return nonzero even if the fs is broken!), which means that
  internal fs errors and corruption are not passed up to userspace
  callers of syncfs(2) or FIFREEZE. Hence fixing the common code and
  XFS, and I'll start working on the ext4/btrfs folks if this is merged.

  Summary:

   - Fix a bug where callers of ->sync_fs (e.g. sync_filesystem and
     syncfs(2)) ignore the return value.

   - Fix a bug where callers of sync_filesystem (e.g. fs freeze) ignore
     the return value.

   - Fix a bug in XFS where xfs_fs_sync_fs never passed back error
     returns"

* tag 'vfs-5.17-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: return errors in xfs_fs_sync_fs
  quota: make dquot_quota_sync return errors from ->sync_fs
  vfs: make sync_filesystem return errors from ->sync_fs
  vfs: make freeze_super abort when sync_filesystem returns error

2 years agoMerge tag 'iomap-5.17-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Linus Torvalds [Sat, 5 Feb 2022 17:04:43 +0000 (09:04 -0800)]
Merge tag 'iomap-5.17-fixes-1' of git://git./fs/xfs/xfs-linux

Pull iomap fix from Darrick Wong:
 "A single bugfix for iomap.

  The fix should eliminate occasional complaints about stall warnings
  when a lot of writeback IO completes all at once and we have to then
  go clearing status on a large number of folios.

  Summary:

   - Limit the length of ioend chains in writeback so that we don't trip
     the softlockup watchdog and to limit long tail latency on clearing
     PageWriteback"

* tag 'iomap-5.17-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs, iomap: limit individual ioend chain lengths in writeback

2 years agoMerge tag 'kvmarm-fixes-5.17-2' of git://git.kernel.org/pub/scm/linux/kernel/git...
Paolo Bonzini [Sat, 5 Feb 2022 05:58:25 +0000 (00:58 -0500)]
Merge tag 'kvmarm-fixes-5.17-2' of git://git./linux/kernel/git/kvmarm/kvmarm into HEAD

KVM/arm64 fixes for 5.17, take #2

- A couple of fixes when handling an exception while a SError has been
  delivered

- Workaround for Cortex-A510's single-step[ erratum

2 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma
Linus Torvalds [Sat, 5 Feb 2022 00:28:11 +0000 (16:28 -0800)]
Merge tag 'for-linus' of git://git./linux/kernel/git/rdma/rdma

Pull rdma fixes from Jason Gunthorpe:
 "Some medium sized bugs in the various drivers. A couple are more
  recent regressions:

   - Fix two panics in hfi1 and two allocation problems

   - Send the IGMP to the correct address in cma

   - Squash a syzkaller bug related to races reading the multicast list

   - Memory leak in siw and cm

   - Fix a corner case spec compliance for HFI/QIB

   - Correct the implementation of fences in siw

   - Error unwind bug in mlx4"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
  RDMA/mlx4: Don't continue event handler after memory allocation failure
  RDMA/siw: Fix broken RDMA Read Fence/Resume logic.
  IB/rdmavt: Validate remote_addr during loopback atomic tests
  IB/cm: Release previously acquired reference counter in the cm_id_priv
  RDMA/siw: Fix refcounting leak in siw_create_qp()
  RDMA/ucma: Protect mc during concurrent multicast leaves
  RDMA/cma: Use correct address when leaving multicast group
  IB/hfi1: Fix tstats alloc and dealloc
  IB/hfi1: Fix AIP early init panic
  IB/hfi1: Fix alloc failure with larger txqueuelen
  IB/hfi1: Fix panic with larger ipoib send_queue_size

2 years agoMerge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Linus Torvalds [Fri, 4 Feb 2022 23:27:45 +0000 (15:27 -0800)]
Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "Seven fixes, six of which are fairly obvious driver fixes.

  The one core change to the device budget depth is to try to ensure
  that if the default depth is large (which can produce quite a sizeable
  bitmap allocation per device), we give back the memory we don't need
  if there's a queue size reduction in slave_configure (which happens to
  a lot of devices)"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: hisi_sas: Fix setting of hisi_sas_slot.is_internal
  scsi: pm8001: Fix use-after-free for aborted SSP/STP sas_task
  scsi: pm8001: Fix use-after-free for aborted TMF sas_task
  scsi: pm8001: Fix warning for undescribed param in process_one_iomb()
  scsi: core: Reallocate device's budget map on queue depth change
  scsi: bnx2fc: Make bnx2fc_recv_frame() mp safe
  scsi: pm80xx: Fix double completion for SATA devices

2 years agoMerge tag 'pci-v5.17-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaa...
Linus Torvalds [Fri, 4 Feb 2022 23:22:35 +0000 (15:22 -0800)]
Merge tag 'pci-v5.17-fixes-3' of git://git./linux/kernel/git/helgaas/pci

Pull pci fixes from Bjorn Helgaas:

 - Restructure j721e_pcie_probe() so we don't dereference a NULL pointer
   (Bjorn Helgaas)

 - Add a kirin_pcie_data struct to identify different Kirin variants to
   fix probe failure for controllers with an internal PHY (Bjorn
   Helgaas)

* tag 'pci-v5.17-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
  PCI: kirin: Add dev struct for of_device_get_match_data()
  PCI: j721e: Initialize pcie->cdns_pcie before using it

2 years agoPCI: kirin: Add dev struct for of_device_get_match_data()
Bjorn Helgaas [Wed, 2 Feb 2022 15:52:41 +0000 (09:52 -0600)]
PCI: kirin: Add dev struct for of_device_get_match_data()

Bean reported that a622435fbe1a ("PCI: kirin: Prefer
of_device_get_match_data()") broke kirin_pcie_probe() because it assumed
match data of 0 was a failure when in fact, it meant the match data was
"(void *)PCIE_KIRIN_INTERNAL_PHY".

Therefore, probing of "hisilicon,kirin960-pcie" devices failed with -EINVAL
and an "OF data missing" message.

Add a struct kirin_pcie_data to encode the PHY type.  Then the result of
of_device_get_match_data() should always be a non-NULL pointer to a struct
kirin_pcie_data that contains the PHY type.

Fixes: a622435fbe1a ("PCI: kirin: Prefer of_device_get_match_data()")
Link: https://lore.kernel.org/r/20220202162659.GA12603@bhelgaas
Link: https://lore.kernel.org/r/20220201215941.1203155-1-huobean@gmail.com
Reported-by: Bean Huo <beanhuo@micron.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2 years agoMerge tag 'for-5.17-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave...
Linus Torvalds [Fri, 4 Feb 2022 20:14:58 +0000 (12:14 -0800)]
Merge tag 'for-5.17-rc2-tag' of git://git./linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:
 "A few fixes and error handling improvements:

   - fix deadlock between quota disable and qgroup rescan worker

   - fix use-after-free after failure to create a snapshot

   - skip warning on unmount after log cleanup failure

   - don't start transaction for scrub if the fs is mounted read-only

   - tree checker verifies item sizes"

* tag 'for-5.17-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: skip reserved bytes warning on unmount after log cleanup failure
  btrfs: fix use of uninitialized variable at rm device ioctl
  btrfs: fix use-after-free after failure to create a snapshot
  btrfs: tree-checker: check item_size for dev_item
  btrfs: tree-checker: check item_size for inode_item
  btrfs: fix deadlock between quota disable and qgroup rescan worker
  btrfs: don't start transaction for scrub if the fs is mounted read-only

2 years agoMerge tag 'erofs-for-5.17-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 4 Feb 2022 20:08:49 +0000 (12:08 -0800)]
Merge tag 'erofs-for-5.17-rc3-fixes' of git://git./linux/kernel/git/xiang/erofs

Pull erofs fixes from Gao Xiang:
 "Two fixes related to fsdax cleanup in this cycle and ztailpacking to
  fix small compressed data inlining. There is also a trivial cleanup to
  rearrange code for better reading.

  Summary:

   - fix fsdax partition offset misbehavior

   - clean up z_erofs_decompressqueue_work() declaration

   - fix up EOF lcluster inlining, especially for small compressed data"

* tag 'erofs-for-5.17-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs:
  erofs: fix small compressed files inlining
  erofs: avoid unnecessary z_erofs_decompressqueue_work() declaration
  erofs: fix fsdax partition offset handling

2 years agoMerge tag 'block-5.17-2022-02-04' of git://git.kernel.dk/linux-block
Linus Torvalds [Fri, 4 Feb 2022 20:01:57 +0000 (12:01 -0800)]
Merge tag 'block-5.17-2022-02-04' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:

 - NVMe pull request
    - fix use-after-free in rdma and tcp controller reset (Sagi Grimberg)
    - fix the state check in nvmf_ctlr_matches_baseopts (Uday Shankar)

 - MD nowait null pointer fix (Song)

 - blk-integrity seed advance fix (Martin)

 - Fix a dio regression in this merge window (Ilya)

* tag 'block-5.17-2022-02-04' of git://git.kernel.dk/linux-block:
  block: bio-integrity: Advance seed correctly for larger interval sizes
  nvme-fabrics: fix state check in nvmf_ctlr_matches_baseopts()
  md: fix NULL pointer deref with nowait but no mddev->queue
  block: fix DIO handling regressions in blkdev_read_iter()
  nvme-rdma: fix possible use-after-free in transport error_recovery work
  nvme-tcp: fix possible use-after-free in transport error_recovery work
  nvme: fix a possible use-after-free in controller reset during load

2 years agoMerge tag 'ata-5.17-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal...
Linus Torvalds [Fri, 4 Feb 2022 19:52:37 +0000 (11:52 -0800)]
Merge tag 'ata-5.17-rc3' of git://git./linux/kernel/git/dlemoal/libata

Pull ATA fixes from Damien Le Moal:

 - Sergey volunteered to be a reviewer for the Renesas R-Car SATA driver
   and PATA drivers. Update the MAINTAINERS file accordingly.

 - Regression fix: add a horkage flag to prevent accessing the log
   directory log page with SATADOM-ML 3ME SATA devices as they react
   badly to reading that log page (from Anton).

* tag 'ata-5.17-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/libata:
  ata: libata-core: Introduce ATA_HORKAGE_NO_LOG_DIR horkage
  MAINTAINERS: add myself as Renesas R-Car SATA driver reviewer
  MAINTAINERS: add myself as PATA drivers reviewer

2 years agoMerge tag 'iommu-fixes-v5.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 4 Feb 2022 19:45:16 +0000 (11:45 -0800)]
Merge tag 'iommu-fixes-v5.17-rc2' of git://git./linux/kernel/git/joro/iommu

Pull iommu fixes from Joerg Roedel:

 - Warning fixes and a fix for a potential use-after-free in IOMMU core
   code

 - Another potential memory leak fix for the Intel VT-d driver

 - Fix for an IO polling loop timeout issue in the AMD IOMMU driver

* tag 'iommu-fixes-v5.17-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
  iommu/amd: Fix loop timeout issue in iommu_ga_log_enable()
  iommu/vt-d: Fix potential memory leak in intel_setup_irq_remapping()
  iommu: Fix some W=1 warnings
  iommu: Fix potential use-after-free during probe

2 years agoMerge tag 'random-5.17-rc3-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 4 Feb 2022 19:38:01 +0000 (11:38 -0800)]
Merge tag 'random-5.17-rc3-for-linus' of git://git./linux/kernel/git/crng/random

Pull random number generator fixes from Jason Donenfeld:
 "For this week, we have:

   - A fix to make more frequent use of hwgenerator randomness, from
     Dominik.

   - More cleanups to the boot initialization sequence, from Dominik.

   - A fix for an old shortcoming with the ZAP ioctl, from me.

   - A workaround for a still unfixed Clang CFI/FullLTO compiler bug,
     from me. On one hand, it's a bummer to commit workarounds for
     experimental compiler features that have bugs. But on the other, I
     think this actually improves the code somewhat, independent of the
     bug. So a win-win"

* tag 'random-5.17-rc3-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/crng/random:
  random: only call crng_finalize_init() for primary_crng
  random: access primary_pool directly rather than through pointer
  random: wake up /dev/random writers after zap
  random: continually use hwgenerator randomness
  lib/crypto: blake2s: avoid indirect calls to compression function for Clang CFI

2 years agoMerge tag 'acpi-5.17-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael...
Linus Torvalds [Fri, 4 Feb 2022 19:32:46 +0000 (11:32 -0800)]
Merge tag 'acpi-5.17-rc3' of git://git./linux/kernel/git/rafael/linux-pm

Pull ACPI fix from Rafael Wysocki:
 "Fix compilation in the case when ACPI is selected and CRC32, depended
  on by ACPI after recent changes, is not (Randy Dunlap)"

* tag 'acpi-5.17-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: require CRC32 to build

2 years agoMerge tag 'sound-5.17-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Fri, 4 Feb 2022 19:24:28 +0000 (11:24 -0800)]
Merge tag 'sound-5.17-rc3' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "A collection of small fixes.

  The major changes are ASoC core fixes, addressing the DPCM locking
  issue after the recent code changes and the potentially invalid
  register accesses via control API. Also, HD-audio got a core fix for
  Oops at dynamic unbinding.

  The rest are device-specific small fixes, including the usual stuff
  like HD-audio and USB-audio quirks"

* tag 'sound-5.17-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (31 commits)
  ALSA: hda: Skip codec shutdown in case the codec is not registered
  ALSA: usb-audio: Correct quirk for VF0770
  ALSA: Replace acpi_bus_get_device()
  Input: wm97xx: Simplify resource management
  ALSA: hda/realtek: Add quirk for ASUS GU603
  ALSA: hda/realtek: Fix silent output on Gigabyte X570 Aorus Xtreme after reboot from Windows
  ALSA: hda/realtek: Fix silent output on Gigabyte X570S Aorus Master (newer chipset)
  ALSA: hda/realtek: Add missing fixup-model entry for Gigabyte X570 ALC1220 quirks
  ALSA: hda: realtek: Fix race at concurrent COEF updates
  ASoC: ops: Check for negative values before reading them
  ASoC: rt5682: Fix deadlock on resume
  ASoC: hdmi-codec: Fix OOB memory accesses
  ASoC: soc-pcm: Move debugfs removal out of spinlock
  ASoC: soc-pcm: Fix DPCM lockdep warning due to nested stream locks
  ASoC: fsl: Add missing error handling in pcm030_fabric_probe
  ALSA: hda: Fix signedness of sscanf() arguments
  ALSA: usb-audio: initialize variables that could ignore errors
  ALSA: hda: Fix UAF of leds class devs at unbinding
  ASoC: qdsp6: q6apm-dai: only stop graphs that are started
  ASoC: codecs: wcd938x: fix return value of mixer put function
  ...

2 years agoMerge tag 'drm-fixes-2022-02-04' of git://anongit.freedesktop.org/drm/drm
Linus Torvalds [Fri, 4 Feb 2022 19:13:54 +0000 (11:13 -0800)]
Merge tag 'drm-fixes-2022-02-04' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "Regular fixes for the week. Daniel has agreed to bring back the fbcon
  hw acceleration under a CONFIG option for the non-drm fbdev users, we
  don't advise turning this on unless you are in the niche that is old
  fbdev drivers, Since it's essentially a revert and shouldn't be high
  impact seemed like a good time to do it now.

  Otherwise, i915 and amdgpu fixes are most of it, along with some minor
  fixes elsewhere.

  fbdev:
   - readd fbcon acceleration

  i915:
   - fix DP monitor via type-c dock
   - fix for engine busyness and read timeout with GuC
   - use ALLOW_FAIL for error capture buffer allocs
   - don't use interruptible lock on error paths
   - smatch fix to reject zero sized overlays.

  amdgpu:
   - mGPU fan boost fix for beige goby
   - S0ix fixes
   - Cyan skillfish hang fix
   - DCN fixes for DCN 3.1
   - DCN fixes for DCN 3.01
   - Apple retina panel fix
   - ttm logic inversion fix

  dma-buf:
   - heaps: fix potential spectre v1 gadget

  kmb:
   - fix potential oob access

  mxsfb:
   - fix NULL ptr deref

  nouveau:
   - fix potential oob access during BIOS decode"

* tag 'drm-fixes-2022-02-04' of git://anongit.freedesktop.org/drm/drm: (24 commits)
  drm: mxsfb: Fix NULL pointer dereference
  drm/amdgpu: fix logic inversion in check
  drm/amd: avoid suspend on dGPUs w/ s2idle support when runtime PM enabled
  drm/amd/display: Force link_rate as LINK_RATE_RBR2 for 2018 15" Apple Retina panels
  drm/amd/display: revert "Reset fifo after enable otg"
  drm/amd/display: watermark latencies is not enough on DCN31
  drm/amd/display: Update watermark values for DCN301
  drm/amdgpu: fix a potential GPU hang on cyan skillfish
  drm/amd: Only run s3 or s0ix if system is configured properly
  drm/amd: add support to check whether the system is set to s3
  fbcon: Add option to enable legacy hardware acceleration
  Revert "fbcon: Disable accelerated scrolling"
  Revert "fbdev: Garbage collect fbdev scrolling acceleration, part 1 (from TODO list)"
  drm/i915/pmu: Fix KMD and GuC race on accessing busyness
  dma-buf: heaps: Fix potential spectre v1 gadget
  drm/amd: Warn users about potential s0ix problems
  drm/amd/pm: correct the MGpuFanBoost support for Beige Goby
  drm/nouveau: fix off by one in BIOS boundary checking
  drm/i915/adlp: Fix TypeC PHY-ready status readout
  drm/i915/pmu: Use PM timestamp instead of RING TIMESTAMP for reference
  ...

2 years agoMerge branch 'akpm' (patches from Andrew)
Linus Torvalds [Fri, 4 Feb 2022 18:34:19 +0000 (10:34 -0800)]
Merge branch 'akpm' (patches from Andrew)

Merge misc fixes from Andrew Morton:
 "10 patches.

  Subsystems affected by this patch series: ipc, MAINTAINERS, and mm
  (vmscan, debug, pagemap, kmemleak, and selftests)"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  kselftest/vm: revert "tools/testing/selftests/vm/userfaultfd.c: use swap() to make code cleaner"
  MAINTAINERS: update rppt's email
  mm/kmemleak: avoid scanning potential huge holes
  ipc/sem: do not sleep with a spin lock held
  mm/pgtable: define pte_index so that preprocessor could recognize it
  mm/page_table_check: check entries at pmd levels
  mm/khugepaged: unify collapse pmd clear, flush and free
  mm/page_table_check: use unsigned long for page counters and cleanup
  mm/debug_vm_pgtable: remove pte entry from the page table
  Revert "mm/page_isolation: unset migratetype directly for non Buddy page"

2 years agorandom: only call crng_finalize_init() for primary_crng
Dominik Brodowski [Sun, 30 Jan 2022 21:03:20 +0000 (22:03 +0100)]
random: only call crng_finalize_init() for primary_crng

crng_finalize_init() returns instantly if it is called for another pool
than primary_crng. The test whether crng_finalize_init() is still required
can be moved to the relevant caller in crng_reseed(), and
crng_need_final_init can be reset to false if crng_finalize_init() is
called with workqueues ready. Then, no previous callsite will call
crng_finalize_init() unless it is needed, and we can get rid of the
superfluous function parameter.

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2 years agorandom: access primary_pool directly rather than through pointer
Dominik Brodowski [Sun, 30 Jan 2022 21:03:19 +0000 (22:03 +0100)]
random: access primary_pool directly rather than through pointer

Both crng_initialize_primary() and crng_init_try_arch_early() are
only called for the primary_pool. Accessing it directly instead of
through a function parameter simplifies the code.

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2 years agorandom: wake up /dev/random writers after zap
Jason A. Donenfeld [Fri, 28 Jan 2022 22:44:03 +0000 (23:44 +0100)]
random: wake up /dev/random writers after zap

When account() is called, and the amount of entropy dips below
random_write_wakeup_bits, we wake up the random writers, so that they
can write some more in. However, the RNDZAPENTCNT/RNDCLEARPOOL ioctl
sets the entropy count to zero -- a potential reduction just like
account() -- but does not unblock writers. This commit adds the missing
logic to that ioctl to unblock waiting writers.

Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2 years agorandom: continually use hwgenerator randomness
Dominik Brodowski [Tue, 25 Jan 2022 20:14:57 +0000 (21:14 +0100)]
random: continually use hwgenerator randomness

The rngd kernel thread may sleep indefinitely if the entropy count is
kept above random_write_wakeup_bits by other entropy sources. To make
best use of multiple sources of randomness, mix entropy from hardware
RNGs into the pool at least once within CRNG_RESEED_INTERVAL.

Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2 years agolib/crypto: blake2s: avoid indirect calls to compression function for Clang CFI
Jason A. Donenfeld [Wed, 19 Jan 2022 13:35:06 +0000 (14:35 +0100)]
lib/crypto: blake2s: avoid indirect calls to compression function for Clang CFI

blake2s_compress_generic is weakly aliased by blake2s_compress. The
current harness for function selection uses a function pointer, which is
ordinarily inlined and resolved at compile time. But when Clang's CFI is
enabled, CFI still triggers when making an indirect call via a weak
symbol. This seems like a bug in Clang's CFI, as though it's bucketing
weak symbols and strong symbols differently. It also only seems to
trigger when "full LTO" mode is used, rather than "thin LTO".

[    0.000000][    T0] Kernel panic - not syncing: CFI failure (target: blake2s_compress_generic+0x0/0x1444)
[    0.000000][    T0] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.16.0-mainline-06981-g076c855b846e #1
[    0.000000][    T0] Hardware name: MT6873 (DT)
[    0.000000][    T0] Call trace:
[    0.000000][    T0]  dump_backtrace+0xfc/0x1dc
[    0.000000][    T0]  dump_stack_lvl+0xa8/0x11c
[    0.000000][    T0]  panic+0x194/0x464
[    0.000000][    T0]  __cfi_check_fail+0x54/0x58
[    0.000000][    T0]  __cfi_slowpath_diag+0x354/0x4b0
[    0.000000][    T0]  blake2s_update+0x14c/0x178
[    0.000000][    T0]  _extract_entropy+0xf4/0x29c
[    0.000000][    T0]  crng_initialize_primary+0x24/0x94
[    0.000000][    T0]  rand_initialize+0x2c/0x6c
[    0.000000][    T0]  start_kernel+0x2f8/0x65c
[    0.000000][    T0]  __primary_switched+0xc4/0x7be4
[    0.000000][    T0] Rebooting in 5 seconds..

Nonetheless, the function pointer method isn't so terrific anyway, so
this patch replaces it with a simple boolean, which also gets inlined
away. This successfully works around the Clang bug.

In general, I'm not too keen on all of the indirection involved here; it
clearly does more harm than good. Hopefully the whole thing can get
cleaned up down the road when lib/crypto is overhauled more
comprehensively. But for now, we go with a simple bandaid.

Fixes: 6048fdcc5f26 ("lib/crypto: blake2s: include as built-in")
Link: https://github.com/ClangBuiltLinux/linux/issues/1567
Reported-by: Miles Chen <miles.chen@mediatek.com>
Tested-by: Miles Chen <miles.chen@mediatek.com>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Tested-by: John Stultz <john.stultz@linaro.org>
Acked-by: Nick Desaulniers <ndesaulniers@google.com>
Reviewed-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2 years agoMerge tag 'ceph-for-5.17-rc3' of git://github.com/ceph/ceph-client
Linus Torvalds [Fri, 4 Feb 2022 17:54:02 +0000 (09:54 -0800)]
Merge tag 'ceph-for-5.17-rc3' of git://github.com/ceph/ceph-client

Pull ceph fixes from Ilya Dryomov:
 "A patch to make it possible to disable zero copy path in the messenger
  to avoid checksum or authentication tag mismatches and ensuing session
  resets in case the destination buffer isn't guaranteed to be stable"

* tag 'ceph-for-5.17-rc3' of git://github.com/ceph/ceph-client:
  libceph: optionally use bounce buffer on recv path in crc mode
  libceph: make recv path in secure mode work the same as send path

2 years agoMerge tag '9p-for-5.17-rc3' of git://github.com/martinetd/linux
Linus Torvalds [Fri, 4 Feb 2022 17:44:42 +0000 (09:44 -0800)]
Merge tag '9p-for-5.17-rc3' of git://github.com/martinetd/linux

Pull 9p fix from Dominique Martinet:
 "Fix 'cannot walk open fid' rule

  The 9p 'walk' operation requires fid arguments to not originate from
  an open or create call and we've missed that for a while as the
  servers regularly running tests with don't enforce the check and no
  active reviewer knew about the rule.

  Both reporters confirmed reverting this patch fixes things for them
  and looking at it further wasn't actually required... Will take more
  time for follow up and enforcing the rule more thoroughly later"

* tag '9p-for-5.17-rc3' of git://github.com/martinetd/linux:
  Revert "fs/9p: search open fids first"

2 years agoMerge tag '5.17-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6
Linus Torvalds [Fri, 4 Feb 2022 17:34:37 +0000 (09:34 -0800)]
Merge tag '5.17-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6

Pull cifs fixes from Steve French:
 "SMB3 client fixes including:

   - multiple fscache related fixes, reenabling ability to read/write to
     cached files for cifs.ko (that was temporarily disabled for cifs.ko
     a few weeks ago due to the recent fscache changes)

   - also includes a new fscache helper function ("query_occupancy")
     used by above

   - fix for multiuser mounts and NTLMSSP auth (workstation name) for
     stable

   - fix locking ordering problem in multichannel code

   - trivial malformed comment fix"

* tag '5.17-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
  cifs: fix workstation_name for multiuser mounts
  Invalidate fscache cookie only when inode attributes are changed.
  cifs: Fix the readahead conversion to manage the batch when reading from cache
  cifs: Implement cache I/O by accessing the cache directly
  netfs, cachefiles: Add a method to query presence of data in the cache
  cifs: Transition from ->readpages() to ->readahead()
  cifs: unlock chan_lock before calling cifs_put_tcp_session
  Fix a warning about a malformed kernel doc comment in cifs

2 years agokselftest/vm: revert "tools/testing/selftests/vm/userfaultfd.c: use swap() to make...
Shuah Khan [Fri, 4 Feb 2022 04:49:45 +0000 (20:49 -0800)]
kselftest/vm: revert "tools/testing/selftests/vm/userfaultfd.c: use swap() to make code cleaner"

With this change, userfaultfd fails to build with undefined reference
swap() error:

  userfaultfd.c: In function `userfaultfd_stress':
  userfaultfd.c:1530:17: warning: implicit declaration of function `swap'; did you mean `swab'? [-Wimplicit-function-declaration]
   1530 |                 swap(area_src, area_dst);
        |                 ^~~~
        |                 swab
  /usr/bin/ld: /tmp/ccDGOAdV.o: in function `userfaultfd_stress':
  userfaultfd.c:(.text+0x549e): undefined reference to `swap'
  /usr/bin/ld: userfaultfd.c:(.text+0x54bc): undefined reference to `swap'
  collect2: error: ld returned 1 exit status

Revert the commit to fix the problem.

Link: https://lkml.kernel.org/r/20220202003340.87195-1-skhan@linuxfoundation.org
Fixes: 2c769ed7137a ("tools/testing/selftests/vm/userfaultfd.c: use swap() to make code cleaner")
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Minghao Chi <chi.minghao@zte.com.cn>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agoMAINTAINERS: update rppt's email
Mike Rapoport [Fri, 4 Feb 2022 04:49:41 +0000 (20:49 -0800)]
MAINTAINERS: update rppt's email

Use my @kernel.org address

Link: https://lkml.kernel.org/r/20220203090324.3701774-1-rppt@kernel.org
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agomm/kmemleak: avoid scanning potential huge holes
Lang Yu [Fri, 4 Feb 2022 04:49:37 +0000 (20:49 -0800)]
mm/kmemleak: avoid scanning potential huge holes

When using devm_request_free_mem_region() and devm_memremap_pages() to
add ZONE_DEVICE memory, if requested free mem region's end pfn were
huge(e.g., 0x400000000), the node_end_pfn() will be also huge (see
move_pfn_range_to_zone()).  Thus it creates a huge hole between
node_start_pfn() and node_end_pfn().

We found on some AMD APUs, amdkfd requested such a free mem region and
created a huge hole.  In such a case, following code snippet was just
doing busy test_bit() looping on the huge hole.

  for (pfn = start_pfn; pfn < end_pfn; pfn++) {
struct page *page = pfn_to_online_page(pfn);
if (!page)
continue;
...
  }

So we got a soft lockup:

  watchdog: BUG: soft lockup - CPU#6 stuck for 26s! [bash:1221]
  CPU: 6 PID: 1221 Comm: bash Not tainted 5.15.0-custom #1
  RIP: 0010:pfn_to_online_page+0x5/0xd0
  Call Trace:
    ? kmemleak_scan+0x16a/0x440
    kmemleak_write+0x306/0x3a0
    ? common_file_perm+0x72/0x170
    full_proxy_write+0x5c/0x90
    vfs_write+0xb9/0x260
    ksys_write+0x67/0xe0
    __x64_sys_write+0x1a/0x20
    do_syscall_64+0x3b/0xc0
    entry_SYSCALL_64_after_hwframe+0x44/0xae

I did some tests with the patch.

(1) amdgpu module unloaded

before the patch:

  real    0m0.976s
  user    0m0.000s
  sys     0m0.968s

after the patch:

  real    0m0.981s
  user    0m0.000s
  sys     0m0.973s

(2) amdgpu module loaded

before the patch:

  real    0m35.365s
  user    0m0.000s
  sys     0m35.354s

after the patch:

  real    0m1.049s
  user    0m0.000s
  sys     0m1.042s

Link: https://lkml.kernel.org/r/20211108140029.721144-1-lang.yu@amd.com
Signed-off-by: Lang Yu <lang.yu@amd.com>
Acked-by: David Hildenbrand <david@redhat.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agoipc/sem: do not sleep with a spin lock held
Minghao Chi [Fri, 4 Feb 2022 04:49:33 +0000 (20:49 -0800)]
ipc/sem: do not sleep with a spin lock held

We can't call kvfree() with a spin lock held, so defer it.

Link: https://lkml.kernel.org/r/20211223031207.556189-1-chi.minghao@zte.com.cn
Fixes: fc37a3b8b438 ("[PATCH] ipc sem: use kvmalloc for sem_undo allocation")
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Reviewed-by: Manfred Spraul <manfred@colorfullife.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Yang Guang <cgel.zte@gmail.com>
Cc: Davidlohr Bueso <dbueso@suse.de>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Bhaskar Chowdhury <unixbhaskar@gmail.com>
Cc: Vasily Averin <vvs@virtuozzo.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agomm/pgtable: define pte_index so that preprocessor could recognize it
Mike Rapoport [Fri, 4 Feb 2022 04:49:29 +0000 (20:49 -0800)]
mm/pgtable: define pte_index so that preprocessor could recognize it

Since commit 974b9b2c68f3 ("mm: consolidate pte_index() and
pte_offset_*() definitions") pte_index is a static inline and there is
no define for it that can be recognized by the preprocessor.  As a
result, vm_insert_pages() uses slower loop over vm_insert_page() instead
of insert_pages() that amortizes the cost of spinlock operations when
inserting multiple pages.

Link: https://lkml.kernel.org/r/20220111145457.20748-1-rppt@kernel.org
Fixes: 974b9b2c68f3 ("mm: consolidate pte_index() and pte_offset_*() definitions")
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Reported-by: Christian Dietrich <stettberger@dokucode.de>
Reviewed-by: Khalid Aziz <khalid.aziz@oracle.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agomm/page_table_check: check entries at pmd levels
Pasha Tatashin [Fri, 4 Feb 2022 04:49:24 +0000 (20:49 -0800)]
mm/page_table_check: check entries at pmd levels

syzbot detected a case where the page table counters were not properly
updated.

  syzkaller login:  ------------[ cut here ]------------
  kernel BUG at mm/page_table_check.c:162!
  invalid opcode: 0000 [#1] PREEMPT SMP KASAN
  CPU: 0 PID: 3099 Comm: pasha Not tainted 5.16.0+ #48
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIO4
  RIP: 0010:__page_table_check_zero+0x159/0x1a0
  Call Trace:
   free_pcp_prepare+0x3be/0xaa0
   free_unref_page+0x1c/0x650
   free_compound_page+0xec/0x130
   free_transhuge_page+0x1be/0x260
   __put_compound_page+0x90/0xd0
   release_pages+0x54c/0x1060
   __pagevec_release+0x7c/0x110
   shmem_undo_range+0x85e/0x1250
  ...

The repro involved having a huge page that is split due to uprobe event
temporarily replacing one of the pages in the huge page.  Later the huge
page was combined again, but the counters were off, as the PTE level was
not properly updated.

Make sure that when PMD is cleared and prior to freeing the level the
PTEs are updated.

Link: https://lkml.kernel.org/r/20220131203249.2832273-5-pasha.tatashin@soleen.com
Fixes: df4e817b7108 ("mm: page table check")
Signed-off-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Slaby <jirislaby@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Paul Turner <pjt@google.com>
Cc: Wei Xu <weixugc@google.com>
Cc: Will Deacon <will@kernel.org>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agomm/khugepaged: unify collapse pmd clear, flush and free
Pasha Tatashin [Fri, 4 Feb 2022 04:49:20 +0000 (20:49 -0800)]
mm/khugepaged: unify collapse pmd clear, flush and free

Unify the code that flushes, clears pmd entry, and frees the PTE table
level into a new function collapse_and_free_pmd().

This cleanup is useful as in the next patch we will add another call to
this function to iterate through PTE prior to freeing the level for page
table check.

Link: https://lkml.kernel.org/r/20220131203249.2832273-4-pasha.tatashin@soleen.com
Signed-off-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Slaby <jirislaby@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Paul Turner <pjt@google.com>
Cc: Wei Xu <weixugc@google.com>
Cc: Will Deacon <will@kernel.org>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agomm/page_table_check: use unsigned long for page counters and cleanup
Pasha Tatashin [Fri, 4 Feb 2022 04:49:15 +0000 (20:49 -0800)]
mm/page_table_check: use unsigned long for page counters and cleanup

For consistency, use "unsigned long" for all page counters.

Also, reduce code duplication by calling __page_table_check_*_clear()
from __page_table_check_*_set() functions.

Link: https://lkml.kernel.org/r/20220131203249.2832273-3-pasha.tatashin@soleen.com
Signed-off-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Reviewed-by: Wei Xu <weixugc@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Slaby <jirislaby@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Paul Turner <pjt@google.com>
Cc: Will Deacon <will@kernel.org>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agomm/debug_vm_pgtable: remove pte entry from the page table
Pasha Tatashin [Fri, 4 Feb 2022 04:49:10 +0000 (20:49 -0800)]
mm/debug_vm_pgtable: remove pte entry from the page table

Patch series "page table check fixes and cleanups", v5.

This patch (of 4):

The pte entry that is used in pte_advanced_tests() is never removed from
the page table at the end of the test.

The issue is detected by page_table_check, to repro compile kernel with
the following configs:

CONFIG_DEBUG_VM_PGTABLE=y
CONFIG_PAGE_TABLE_CHECK=y
CONFIG_PAGE_TABLE_CHECK_ENFORCED=y

During the boot the following BUG is printed:

  debug_vm_pgtable: [debug_vm_pgtable         ]: Validating architecture page table helpers
  ------------[ cut here ]------------
  kernel BUG at mm/page_table_check.c:162!
  invalid opcode: 0000 [#1] PREEMPT SMP PTI
  CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.16.0-11413-g2c271fe77d52 #3
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.15.0-0-g2dd4b9b3f840-prebuilt.qemu.org 04/01/2014
  ...

The entry should be properly removed from the page table before the page
is released to the free list.

Link: https://lkml.kernel.org/r/20220131203249.2832273-1-pasha.tatashin@soleen.com
Link: https://lkml.kernel.org/r/20220131203249.2832273-2-pasha.tatashin@soleen.com
Fixes: a5c3b9ffb0f4 ("mm/debug_vm_pgtable: add tests validating advanced arch page table helpers")
Signed-off-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Tested-by: Zi Yan <ziy@nvidia.com>
Acked-by: David Rientjes <rientjes@google.com>
Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Wei Xu <weixugc@google.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Will Deacon <will@kernel.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Jiri Slaby <jirislaby@kernel.org>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: <stable@vger.kernel.org> [5.9+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agoRevert "mm/page_isolation: unset migratetype directly for non Buddy page"
Chen Wandun [Fri, 4 Feb 2022 04:49:06 +0000 (20:49 -0800)]
Revert "mm/page_isolation: unset migratetype directly for non Buddy page"

This reverts commit 721fb891ad0b3956d5c168b2931e3e5e4fb7ca40.

Commit 721fb891ad0b ("mm/page_isolation: unset migratetype directly for
non Buddy page") will result memory that should in buddy disappear by
mistake.  move_freepages_block moves all pages in pageblock instead of
pages indicated by input parameter, so if input pages is not in buddy
but other pages in pageblock is in buddy, it will result in page out of
control.

Link: https://lkml.kernel.org/r/20220126024436.13921-1-chenwandun@huawei.com
Fixes: 721fb891ad0b ("mm/page_isolation: unset migratetype directly for non Buddy page")
Signed-off-by: Chen Wandun <chenwandun@huawei.com>
Reported-by: "kernelci.org bot" <bot@kernelci.org>
Acked-by: David Hildenbrand <david@redhat.com>
Tested-by: Dong Aisheng <aisheng.dong@nxp.com>
Tested-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2 years agoiommu/amd: Fix loop timeout issue in iommu_ga_log_enable()
Joerg Roedel [Fri, 4 Feb 2022 11:55:37 +0000 (12:55 +0100)]
iommu/amd: Fix loop timeout issue in iommu_ga_log_enable()

The polling loop for the register change in iommu_ga_log_enable() needs
to have a udelay() in it.  Otherwise the CPU might be faster than the
IOMMU hardware and wrongly trigger the WARN_ON() further down the code
stream. Use a 10us for udelay(), has there is some hardware where
activation of the GA log can take more than a 100ms.

A future optimization should move the activation check of the GA log
to the point where it gets used for the first time. But that is a
bigger change and not suitable for a fix.

Fixes: 8bda0cfbdc1a ("iommu/amd: Detect and initialize guest vAPIC log")
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Link: https://lore.kernel.org/r/20220204115537.3894-1-joro@8bytes.org
2 years agoPCI/MSI: Remove bogus warning in pci_irq_get_affinity()
Thomas Gleixner [Mon, 31 Jan 2022 21:02:46 +0000 (22:02 +0100)]
PCI/MSI: Remove bogus warning in pci_irq_get_affinity()

The recent overhaul of pci_irq_get_affinity() introduced a regression when
pci_irq_get_affinity() is called for an MSI-X interrupt which was not
allocated with affinity descriptor information.

The original code just returned a NULL pointer in that case, but the rework
added a WARN_ON() under the assumption that the corresponding WARN_ON() in
the MSI case can be applied to MSI-X as well.

In fact the MSI warning in the original code does not make sense either
because it's legitimate to invoke pci_irq_get_affinity() for a MSI
interrupt which was not allocated with affinity descriptor information.

Remove it and just return NULL as the original code did.

Fixes: f48235900182 ("PCI/MSI: Simplify pci_irq_get_affinity()")
Reported-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/87ee4n38sm.ffs@tglx
2 years agoKVM: x86: Use ERR_PTR_USR() to return -EFAULT as a __user pointer
Sean Christopherson [Wed, 2 Feb 2022 00:51:57 +0000 (00:51 +0000)]
KVM: x86: Use ERR_PTR_USR() to return -EFAULT as a __user pointer

Use ERR_PTR_USR() when returning -EFAULT from kvm_get_attr_addr(), sparse
complains about implicitly casting the kernel pointer from ERR_PTR() into
a __user pointer.

>> arch/x86/kvm/x86.c:4342:31: sparse: sparse: incorrect type in return expression
   (different address spaces) @@     expected void [noderef] __user * @@     got void * @@
   arch/x86/kvm/x86.c:4342:31: sparse:     expected void [noderef] __user *
   arch/x86/kvm/x86.c:4342:31: sparse:     got void *
>> arch/x86/kvm/x86.c:4342:31: sparse: sparse: incorrect type in return expression
   (different address spaces) @@     expected void [noderef] __user * @@     got void * @@
   arch/x86/kvm/x86.c:4342:31: sparse:     expected void [noderef] __user *
   arch/x86/kvm/x86.c:4342:31: sparse:     got void *

No functional change intended.

Fixes: 56f289a8d23a ("KVM: x86: Add a helper to retrieve userspace address from kvm_device_attr")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220202005157.2545816-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoKVM: x86: Report deprecated x87 features in supported CPUID
Jim Mattson [Fri, 4 Feb 2022 00:13:48 +0000 (16:13 -0800)]
KVM: x86: Report deprecated x87 features in supported CPUID

CPUID.(EAX=7,ECX=0):EBX.FDP_EXCPTN_ONLY[bit 6] and
CPUID.(EAX=7,ECX=0):EBX.ZERO_FCS_FDS[bit 13] are "defeature"
bits. Unlike most of the other CPUID feature bits, these bits are
clear if the features are present and set if the features are not
present. These bits should be reported in KVM_GET_SUPPORTED_CPUID,
because if these bits are set on hardware, they cannot be cleared in
the guest CPUID. Doing so would claim guest support for a feature that
the hardware doesn't support and that can't be efficiently emulated.

Of course, any software (e.g WIN87EM.DLL) expecting these features to
be present likely predates these CPUID feature bits and therefore
doesn't know to check for them anyway.

Aaron Lewis added the corresponding X86_FEATURE macros in
commit cbb99c0f5887 ("x86/cpufeatures: Add FDP_EXCPTN_ONLY and
ZERO_FCS_FDS"), with the intention of reporting these bits in
KVM_GET_SUPPORTED_CPUID, but I was unable to find a proposed patch on
the kvm list.

Opportunistically reordered the CPUID_7_0_EBX capability bits from
least to most significant.

Cc: Aaron Lewis <aaronlewis@google.com>
Signed-off-by: Jim Mattson <jmattson@google.com>
Message-Id: <20220204001348.2844660-1-jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoata: libata-core: Introduce ATA_HORKAGE_NO_LOG_DIR horkage
Anton Lundin [Thu, 3 Feb 2022 09:41:35 +0000 (10:41 +0100)]
ata: libata-core: Introduce ATA_HORKAGE_NO_LOG_DIR horkage

06f6c4c6c3e8 ("ata: libata: add missing ata_identify_page_supported() calls")
introduced additional calls to ata_identify_page_supported(), thus also
adding indirectly accesses to the device log directory log page through
ata_log_supported(). Reading this log page causes SATADOM-ML 3ME devices
to lock up.

Introduce the horkage flag ATA_HORKAGE_NO_LOG_DIR to prevent accesses to
the log directory in ata_log_supported() and add a blacklist entry
with this flag for "SATADOM-ML 3ME" devices.

Fixes: 636f6e2af4fb ("libata: add horkage for missing Identify Device log")
Cc: stable@vger.kernel.org # v5.10+
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
2 years agoMAINTAINERS: add myself as Renesas R-Car SATA driver reviewer
Sergey Shtylyov [Thu, 3 Feb 2022 19:47:09 +0000 (22:47 +0300)]
MAINTAINERS: add myself as Renesas R-Car SATA driver reviewer

Add myself as a reviewer for the Renesas R-Car SATA driver -- I don't have
the hardware anymore (Geert Uytterhoeven does have a lot of hardware!) but
I do have the manuals still! :-)

Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
2 years agoMerge tag 'drm-intel-fixes-2022-02-03' of git://anongit.freedesktop.org/drm/drm-intel...
Dave Airlie [Fri, 4 Feb 2022 05:48:26 +0000 (15:48 +1000)]
Merge tag 'drm-intel-fixes-2022-02-03' of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes

Fix GitLab issue #4698: DP monitor through Type-C dock(Dell DA310) doesn't work.
Fixes for inconsistent engine busyness value and read timeout with GuC.
Fix to use ALLOW_FAIL for error capture buffer allocation. Don't use
interruptible lock on error path. Smatch fix to reject zero sized overlays.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/YfuiG8SKMKP5V/Dm@jlahtine-mobl.ger.corp.intel.com