linux-stable.git/drivers/clocksource/hyperv_timer.c, branch linux-5.10.y

x86/hyperv: Fix hv tsc page based sched_clock for hibernation

2025-01-09T12:25:00+00:00

commit bcc80dec91ee745b3d66f3e48f0ec2efdea97149 upstream.

read_hv_sched_clock_tsc() assumes that the Hyper-V clock counter is
bigger than the variable hv_sched_clock_offset, which is cached during
early boot, but depending on the timing this assumption may be false
when a hibernated VM starts again (the clock counter starts from 0
again) and is resuming back (Note: hv_init_tsc_clocksource() is not
called during hibernation/resume); consequently,
read_hv_sched_clock_tsc() may return a negative integer (which is
interpreted as a huge positive integer since the return type is u64)
and new kernel messages are prefixed with huge timestamps before
read_hv_sched_clock_tsc() grows big enough (which typically takes
several seconds).

Fix the issue by saving the Hyper-V clock counter just before the
suspend, and using it to correct the hv_sched_clock_offset in
resume. This makes hv tsc page based sched_clock continuous and ensures
that post resume, it starts from where it left off during suspend.
Override x86_platform.save_sched_clock_state and
x86_platform.restore_sched_clock_state routines to correct this as soon
as possible.

Note: if Invariant TSC is available, the issue doesn't happen because
1) we don't register read_hv_sched_clock_tsc() for sched clock:
See commit e5313f1c5404 ("clocksource/drivers/hyper-v: Rework
clocksource and sched clock setup");
2) the common x86 code adjusts TSC similarly: see
__restore_processor_state() ->  tsc_verify_tsc_adjust(true) and
x86_platform.restore_sched_clock_state().

Cc: stable@vger.kernel.org
Fixes: 1349401ff1aa ("clocksource/drivers/hyper-v: Suspend/resume Hyper-V clocksource for hibernation")
Co-developed-by: Dexuan Cui 
Signed-off-by: Dexuan Cui 
Signed-off-by: Naman Jain 
Reviewed-by: Michael Kelley 
Link: https://lore.kernel.org/r/20240917053917.76787-1-namjain@linux.microsoft.com
Signed-off-by: Wei Liu 
Message-ID: <20240917053917.76787-1-namjain@linux.microsoft.com>
Signed-off-by: Naman Jain 
Signed-off-by: Greg Kroah-Hartman

clocksource: hyper-v: unexport __init-annotated hv_init_clocksource()

2022-06-22T12:13:16+00:00

[ Upstream commit 245b993d8f6c4e25f19191edfbd8080b645e12b1 ]

EXPORT_SYMBOL and __init is a bad combination because the .init.text
section is freed up after the initialization. Hence, modules cannot
use symbols annotated __init. The access to a freed symbol may end up
with kernel panic.

modpost used to detect it, but it has been broken for a decade.

Recently, I fixed modpost so it started to warn it again, then this
showed up in linux-next builds.

There are two ways to fix it:

  - Remove __init
  - Remove EXPORT_SYMBOL

I chose the latter for this case because the only in-tree call-site,
arch/x86/kernel/cpu/mshyperv.c is never compiled as modular.
(CONFIG_HYPERVISOR_GUEST is boolean)

Fixes: dd2cb348613b ("clocksource/drivers: Continue making Hyper-V clocksource ISA agnostic")
Reported-by: Stephen Rothwell 
Signed-off-by: Masahiro Yamada 
Reviewed-by: Vitaly Kuznetsov 
Reviewed-by: Michael Kelley 
Link: https://lore.kernel.org/r/20220606050238.4162200-1-masahiroy@kernel.org
Signed-off-by: Wei Liu 
Signed-off-by: Sasha Levin

hv: clocksource: Add notrace attribute to read_hv_sched_clock_*() functions

2020-09-28T09:04:48+00:00

When selecting function_graph tracer with the command:
 # echo function_graph > /sys/kernel/debug/tracing/current_tracer

The kernel crashes with the following stack trace:

[69703.122389] BUG: stack guard page was hit at 000000001056545c (stack is 00000000fa3f8fed..0000000005d39503)
[69703.122403] kernel stack overflow (double-fault): 0000 [#1] SMP PTI
[69703.122413] CPU: 0 PID: 16982 Comm: bash Kdump: loaded Not tainted 4.18.0-236.el8.x86_64 #1
[69703.122420] Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS Hyper-V UEFI Release v4.0 12/17/2019
[69703.122433] RIP: 0010repare_ftrace_return+0xa/0x110
[69703.122458] Code: 05 00 0f 0b 48 c7 c7 10 ca 69 ae 0f b6 f0 e8 4b 52 0c 00 31 c0 eb ca 66 0f 1f 84 00 00 00 00 00 55 48 89 e5 41 56 41 55 41 54 <53> 48 83 ec 18 65 48 8b 04 25 28 00 00 00 48 89 45 d8 31 c0 48 85
[69703.122467] RSP: 0018:ffffbd6d01118000 EFLAGS: 00010086
[69703.122476] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000003
[69703.122484] RDX: 0000000000000000 RSI: ffffbd6d011180d8 RDI: ffffffffadce7550
[69703.122491] RBP: ffffbd6d01118018 R08: 0000000000000000 R09: ffff9d4b09266000
[69703.122498] R10: ffff9d4b0fc04540 R11: ffff9d4b0fc20a00 R12: ffff9d4b6e42aa90
[69703.122506] R13: ffff9d4b0fc20ab8 R14: 00000000000003e8 R15: ffffbd6d0111837c
[69703.122514] FS:  00007fd5f2588740(0000) GS:ffff9d4b6e400000(0000) knlGS:0000000000000000
[69703.122521] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[69703.122528] CR2: ffffbd6d01117ff8 CR3: 00000000565d8001 CR4: 00000000003606f0
[69703.122538] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[69703.122545] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[69703.122552] Call Trace:
[69703.122568]  ftrace_graph_caller+0x6b/0xa0
[69703.122589]  ? read_hv_sched_clock_tsc+0x5/0x20
[69703.122599]  read_hv_sched_clock_tsc+0x5/0x20
[69703.122611]  sched_clock+0x5/0x10
[69703.122621]  sched_clock_local+0x12/0x80
[69703.122631]  sched_clock_cpu+0x8c/0xb0
[69703.122644]  trace_clock_global+0x21/0x90
[69703.122655]  ring_buffer_lock_reserve+0x100/0x3c0
[69703.122671]  trace_buffer_lock_reserve+0x16/0x50
[69703.122683]  __trace_graph_entry+0x28/0x90
[69703.122695]  trace_graph_entry+0xfd/0x1a0
[69703.122705]  ? read_hv_clock_tsc_cs+0x10/0x10
[69703.122714]  ? sched_clock+0x5/0x10
[69703.122723]  prepare_ftrace_return+0x99/0x110
[69703.122734]  ? read_hv_clock_tsc_cs+0x10/0x10
[69703.122743]  ? sched_clock+0x5/0x10
[69703.122752]  ftrace_graph_caller+0x6b/0xa0
[69703.122768]  ? read_hv_clock_tsc_cs+0x10/0x10
[69703.122777]  ? sched_clock+0x5/0x10
[69703.122786]  ? read_hv_sched_clock_tsc+0x5/0x20
[69703.122796]  ? ring_buffer_unlock_commit+0x1d/0xa0
[69703.122805]  read_hv_sched_clock_tsc+0x5/0x20
[69703.122814]  ftrace_graph_caller+0xa0/0xa0
[ ... recursion snipped ... ]

Setting the notrace attribute for read_hv_sched_clock_msr() and
read_hv_sched_clock_tsc() fixes it.

Fixes: bd00cd52d5be ("clocksource/drivers/hyperv: Add Hyper-V specific sched clock function")
Signed-off-by: Vitaly Kuznetsov 
Signed-off-by: Mohammed Gamal 
Link: https://lore.kernel.org/r/20200924151117.767442-1-mgamal@redhat.com
Signed-off-by: Wei Liu

Merge tag 'timers-core-2020-03-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2020-03-31T01:51:47+00:00

Pull timekeeping and timer updates from Thomas Gleixner:
 "Core:

   - Consolidation of the vDSO build infrastructure to address the
     difficulties of cross-builds for ARM64 compat vDSO libraries by
     restricting the exposure of header content to the vDSO build.

     This is achieved by splitting out header content into separate
     headers. which contain only the minimaly required information which
     is necessary to build the vDSO. These new headers are included from
     the kernel headers and the vDSO specific files.

   - Enhancements to the generic vDSO library allowing more fine grained
     control over the compiled in code, further reducing architecture
     specific storage and preparing for adopting the generic library by
     PPC.

   - Cleanup and consolidation of the exit related code in posix CPU
     timers.

   - Small cleanups and enhancements here and there

  Drivers:

   - The obligatory new drivers: Ingenic JZ47xx and X1000 TCU support

   - Correct the clock rate of PIT64b global clock

   - setup_irq() cleanup

   - Preparation for PWM and suspend support for the TI DM timer

   - Expand the fttmr010 driver to support ast2600 systems

   - The usual small fixes, enhancements and cleanups all over the
     place"

* tag 'timers-core-2020-03-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (80 commits)
  Revert "clocksource/drivers/timer-probe: Avoid creating dead devices"
  vdso: Fix clocksource.h macro detection
  um: Fix header inclusion
  arm64: vdso32: Enable Clang Compilation
  lib/vdso: Enable common headers
  arm: vdso: Enable arm to use common headers
  x86/vdso: Enable x86 to use common headers
  mips: vdso: Enable mips to use common headers
  arm64: vdso32: Include common headers in the vdso library
  arm64: vdso: Include common headers in the vdso library
  arm64: Introduce asm/vdso/processor.h
  arm64: vdso32: Code clean up
  linux/elfnote.h: Replace elf.h with UAPI equivalent
  scripts: Fix the inclusion order in modpost
  common: Introduce processor.h
  linux/ktime.h: Extract common header for vDSO
  linux/jiffies.h: Extract common header for vDSO
  linux/time64.h: Extract common header for vDSO
  linux/time32.h: Extract common header for vDSO
  linux/time.h: Extract common header for vDSO
  ...

clocksource/drivers/hyper-v: Make sched clock return nanoseconds correctly

2020-03-27T11:27:45+00:00

The sched clock read functions return the HV clock (100ns granularity)
without converting it to nanoseconds.

Add the missing conversion.

Fixes: bd00cd52d5be ("clocksource/drivers/hyperv: Add Hyper-V specific sched clock function")
Signed-off-by: Yubo Xie 
Signed-off-by: Tianyu Lan 
Signed-off-by: Thomas Gleixner 
Reviewed-by: Vitaly Kuznetsov 
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20200327021159.31429-1-Tianyu.Lan@microsoft.com

x86/vdso: Move VDSO clocksource state tracking to callback

2020-02-17T13:40:22+00:00

All architectures which use the generic VDSO code have their own storage
for the VDSO clock mode. That's pointless and just requires duplicate code.

X86 abuses the function which retrieves the architecture specific clock
mode storage to mark the clocksource as used in the VDSO. That's silly
because this is invoked on every tick when the VDSO data is updated.

Move this functionality to the clocksource::enable() callback so it gets
invoked once when the clocksource is installed. This allows to make the
clock mode storage generic.

Signed-off-by: Thomas Gleixner 
Reviewed-by: Michael Kelley   (Hyper-V parts)
Reviewed-by: Vincenzo Frascino  (VDSO parts)
Acked-by: Juergen Gross  (Xen parts)
Link: https://lkml.kernel.org/r/20200207124402.934519777@linutronix.de

clocksource/drivers/hyper-v: Set TSC clocksource as default w/ InvariantTSC

2020-01-16T18:09:02+00:00

Change the Hyper-V clocksource ratings to 250, below the TSC clocksource
rating of 300.  In configurations where Hyper-V offers an InvariantTSC,
the TSC is not marked "unstable", so the TSC clocksource is available
and preferred.  With the higher rating, it will be the default.  On
older hardware and Hyper-V versions, the TSC is marked "unstable", so no
TSC clocksource is created and the selected Hyper-V clocksource will be
the default.

Signed-off-by: Andrea Parri 
Reviewed-by: Michael Kelley 
Signed-off-by: Daniel Lezcano 
Link: https://lore.kernel.org/r/20200109160650.16150-3-parri.andrea@gmail.com

clocksource/drivers/hyper-v: Untangle stimers and timesync from clocksources

2020-01-16T18:09:02+00:00

hyperv_timer.c exports hyperv_cs, which is used by stimers and the
timesync mechanism.  However, the clocksource dependency is not
needed: these mechanisms only depend on the partition reference
counter (which can be read via a MSR or via the TSC Reference Page).

Introduce the (function) pointer hv_read_reference_counter, as an
embodiment of the partition reference counter read, and export it
in place of the hyperv_cs pointer.  The latter can be removed.

This should clarify that there's no relationship between Hyper-V
stimers & timesync and the Linux clocksource abstractions.  No
functional or semantic change.

Suggested-by: Michael Kelley 
Signed-off-by: Andrea Parri 
Reviewed-by: Michael Kelley 
Signed-off-by: Daniel Lezcano 
Link: https://lore.kernel.org/r/20200109160650.16150-2-parri.andrea@gmail.com

clocksource/drivers/hyper-v: Reserve PAGE_SIZE space for tsc page

2020-01-16T18:07:09+00:00

Currently, the reserved size for a tsc page is 4K, which is enough for
communicating with hypervisor. However, in the case where we want to
export the tsc page to userspace (e.g. for vDSO to read the
clocksource), the tsc page should be at least PAGE_SIZE, otherwise, when
PAGE_SIZE is larger than 4K, extra kernel data will be mapped into
userspace, which means leaking kernel information.

Therefore reserve PAGE_SIZE space for tsc_pg as a preparation for the
vDSO support of ARM64 in the future. Also, while at it, replace all
reference to tsc_pg with hv_get_tsc_page() since it should be the only
interface to access tsc page.

Signed-off-by: Boqun Feng (Microsoft) 
Cc: linux-hyperv@vger.kernel.org
Reviewed-by: Michael Kelley 
Signed-off-by: Daniel Lezcano 
Link: https://lore.kernel.org/r/20191126021723.4710-1-boqun.feng@gmail.com

clocksource/drivers/hyper-v: Suspend/resume Hyper-V clocksource for hibernation

2020-01-16T18:06:57+00:00

This is needed for hibernation, e.g. when we resume the old kernel, we need
to disable the "current" kernel's TSC page and then resume the old kernel's.

Signed-off-by: Dexuan Cui 
Reviewed-by: Michael Kelley 
Signed-off-by: Daniel Lezcano 
Link: https://lore.kernel.org/r/1574233946-48377-1-git-send-email-decui@microsoft.com