[PATCH v2] powerpc: Avoid nmi_enter/nmi_exit in real mode interrupt.

[Christophe Leroy]

Le 05/02/2024 à 06:36, Mahesh Salgaonkar a écrit :
> [Vous ne recevez pas souvent de courriers de mahesh at linux.ibm.com. Découvrez pourquoi ceci est important à https://aka.ms/LearnAboutSenderIdentification ]
> 
> nmi_enter()/nmi_exit() touches per cpu variables which can lead to kernel
> crash when invoked during real mode interrupt handling (e.g. early HMI/MCE
> interrupt handler) if percpu allocation comes from vmalloc area.
> 
> Early HMI/MCE handlers are called through DEFINE_INTERRUPT_HANDLER_NMI()
> wrapper which invokes nmi_enter/nmi_exit calls. We don't see any issue when
> percpu allocation is from the embedded first chunk. However with
> CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK enabled there are chances where percpu
> allocation can come from the vmalloc area.
> 
> With kernel command line "percpu_alloc=page" we can force percpu allocation
> to come from vmalloc area and can see kernel crash in machine_check_early:
> 
> [    1.215714] NIP [c000000000e49eb4] rcu_nmi_enter+0x24/0x110
> [    1.215717] LR [c0000000000461a0] machine_check_early+0xf0/0x2c0
> [    1.215719] --- interrupt: 200
> [    1.215720] [c000000fffd73180] [0000000000000000] 0x0 (unreliable)
> [    1.215722] [c000000fffd731b0] [0000000000000000] 0x0
> [    1.215724] [c000000fffd73210] [c000000000008364] machine_check_early_common+0x134/0x1f8
> 
> Fix this by avoiding use of nmi_enter()/nmi_exit() in real mode if percpu
> first chunk is not embedded.
> 
> Signed-off-by: Mahesh Salgaonkar <mahesh at linux.ibm.com>
> ---
> Changes in v2:
> - Rebase to upstream master
> - Use jump_labels, if CONFIG_JUMP_LABEL is enabled, to avoid redoing the
>    test at each interrupt entry.
> - v1 is at https://lore.kernel.org/linuxppc-dev/164578465828.74956.6065296024817333750.stgit@jupiter/
> ---
>   arch/powerpc/include/asm/interrupt.h | 14 ++++++++++++++
>   arch/powerpc/include/asm/percpu.h    | 11 +++++++++++
>   arch/powerpc/kernel/setup_64.c       | 12 ++++++++++++
>   3 files changed, 37 insertions(+)
> 
> diff --git a/arch/powerpc/include/asm/interrupt.h b/arch/powerpc/include/asm/interrupt.h
> index a4196ab1d0167..3b4e17c23d9a9 100644
> --- a/arch/powerpc/include/asm/interrupt.h
> +++ b/arch/powerpc/include/asm/interrupt.h
> @@ -336,6 +336,16 @@ static inline void interrupt_nmi_enter_prepare(struct pt_regs *regs, struct inte
>          if (IS_ENABLED(CONFIG_KASAN))
>                  return;
> 
> +       /*
> +        * Likewise, do not use it in real mode if percpu first chunk is not
> +        * embedded. With CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK enabled there
> +        * are chances where percpu allocation can come from vmalloc area.
> +        */
> +#ifdef CONFIG_PPC64

Instead of adding this #ifdef in middle of code, could you define 
is_embed_first_chunk as always 'true' when CONFIG_PPC64 is not defined ?

> +       if (IS_ENABLED(CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK) && !is_embed_first_chunk)
> +               return;
> +#endif
> +
>          /* Otherwise, it should be safe to call it */
>          nmi_enter();
>   }
> @@ -351,6 +361,10 @@ static inline void interrupt_nmi_exit_prepare(struct pt_regs *regs, struct inter
>                  // no nmi_exit for a pseries hash guest taking a real mode exception
>          } else if (IS_ENABLED(CONFIG_KASAN)) {
>                  // no nmi_exit for KASAN in real mode
> +#ifdef CONFIG_PPC64

Same

> +       } else if (IS_ENABLED(CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK) && !is_embed_first_chunk) {
> +               // no nmi_exit if percpu first chunk is not embedded
> +#endif
>          } else {
>                  nmi_exit();
>          }
> diff --git a/arch/powerpc/include/asm/percpu.h b/arch/powerpc/include/asm/percpu.h
> index 8e5b7d0b851c6..6b4dce4e78d5f 100644
> --- a/arch/powerpc/include/asm/percpu.h
> +++ b/arch/powerpc/include/asm/percpu.h
> @@ -12,6 +12,17 @@
> 
>   #define __my_cpu_offset local_paca->data_offset
> 
> +#ifdef CONFIG_JUMP_LABEL
> +DECLARE_STATIC_KEY_FALSE(__percpu_embed_first_chunk);
> +
> +#define is_embed_first_chunk   \
> +               (static_key_enabled(&__percpu_embed_first_chunk.key))
> +
> +#else /* !CONFIG_JUMP_LABEL */
> +extern bool __percpu_embed_first_chunk;
> +#define is_embed_first_chunk   __percpu_embed_first_chunk
> +
> +#endif /* CONFIG_JUMP_LABEL */
>   #endif /* CONFIG_SMP */
>   #endif /* __powerpc64__ */
> 
> diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
> index 2f19d5e944852..674b6e1bebe9a 100644
> --- a/arch/powerpc/kernel/setup_64.c
> +++ b/arch/powerpc/kernel/setup_64.c
> @@ -834,6 +834,11 @@ static __init int pcpu_cpu_to_node(int cpu)
> 
>   unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
>   EXPORT_SYMBOL(__per_cpu_offset);
> +#ifdef CONFIG_JUMP_LABEL

Why this ifdef ? Even when CONFIG_JUMP_LABEL is not selected all this 
should just work fine.

> +DEFINE_STATIC_KEY_FALSE(__percpu_embed_first_chunk);
> +#else
> +bool __percpu_embed_first_chunk;
> +#endif
> 
>   void __init setup_per_cpu_areas(void)
>   {
> @@ -869,6 +874,13 @@ void __init setup_per_cpu_areas(void)
>                          pr_warn("PERCPU: %s allocator failed (%d), "
>                                  "falling back to page size\n",
>                                  pcpu_fc_names[pcpu_chosen_fc], rc);
> +               else {
> +#ifdef CONFIG_JUMP_LABEL
> +                       static_key_enable(&__percpu_embed_first_chunk.key);
> +#else
> +                       __percpu_embed_first_chunk = true;
> +#endif
> +               }
>          }
> 
>          if (rc < 0)
> --
> 2.43.0
>