[PATCH v4 11/18] KVM: arm64: Stub implementation of non-protected nVHE HYP stack unwinder

Marc Zyngier maz at kernel.org
Mon Jul 18 09:57:25 PDT 2022


On 2022-07-18 17:51, Kalesh Singh wrote:
> On Mon, Jul 18, 2022 at 12:31 AM Marc Zyngier <maz at kernel.org> wrote:
>> 
>> On Fri, 15 Jul 2022 07:10:20 +0100,
>> Kalesh Singh <kaleshsingh at google.com> wrote:
>> >
>> > Add stub implementations of non-protected nVHE stack unwinder, for
>> > building. These are implemented later in this series.
>> >
>> > Signed-off-by: Kalesh Singh <kaleshsingh at google.com>
>> > ---
>> >  arch/arm64/include/asm/stacktrace/nvhe.h | 22 ++++++++++++++++++++++
>> >  1 file changed, 22 insertions(+)
>> >
>> > diff --git a/arch/arm64/include/asm/stacktrace/nvhe.h b/arch/arm64/include/asm/stacktrace/nvhe.h
>> > index 1eac4e57f2ae..36cf7858ddd8 100644
>> > --- a/arch/arm64/include/asm/stacktrace/nvhe.h
>> > +++ b/arch/arm64/include/asm/stacktrace/nvhe.h
>> > @@ -8,6 +8,12 @@
>> >   *      the HYP memory. The stack is unwinded in EL2 and dumped to a shared
>> >   *      buffer where the host can read and print the stacktrace.
>> >   *
>> > + *   2) Non-protected nVHE mode - the host can directly access the
>> > + *      HYP stack pages and unwind the HYP stack in EL1. This saves having
>> > + *      to allocate shared buffers for the host to read the unwinded
>> > + *      stacktrace.
>> > + *
>> > + *
>> >   * Copyright (C) 2022 Google LLC
>> >   */
>> >  #ifndef __ASM_STACKTRACE_NVHE_H
>> > @@ -53,5 +59,21 @@ static int notrace unwind_next(struct unwind_state *state)
>> >  NOKPROBE_SYMBOL(unwind_next);
>> >  #endif       /* CONFIG_PROTECTED_NVHE_STACKTRACE */
>> >
>> > +/*
>> > + * Non-protected nVHE HYP stack unwinder
>> > + */
>> > +#else        /* !__KVM_NVHE_HYPERVISOR__ */
>> 
>> I don't get this path. This either represents the VHE hypervisor or
>> the kernel proper. Which one is it?
> 
> Hi Marc. This is run from kernel proper context. And it's the
> unwinding for conventional nVHE (non-protected). The unwinding is done
> from the host kernel in EL1.

This really deserves a comment here, as the one you currently
have is a bit misleading (and you probably want to move it
below the #else).

Thanks,

         M.
-- 
Jazz is not dead. It just smells funny...



More information about the linux-arm-kernel mailing list