[PATCH] perf/core: Avoid freeing static PMU contexts when PMU is unregistered

[Mark Rutland]

On Tue, Oct 03, 2017 at 03:20:50PM +0100, Will Deacon wrote:
> Since commit 1fd7e4169954 ("perf/core: Remove perf_cpu_context::unique_pmu"),
> when a PMU is unregistered then its associated ->pmu_cpu_context is
> unconditionally freed. Whilst this is fine for dynamically allocated
> context types (i.e. those registered using perf_invalid_context), this
> causes a problem for sharing of static contexts such as
> perf_{sw,hw}_context, which are used by multiple built-in PMUs and
> effectively have a global lifetime.
> 
> Whilst testing the ARM SPE driver, which must use perf_sw_context to
> support per-task AUX tracing, unregistering the driver as a result of a
> module unload resulted in:
> 
>  Unable to handle kernel NULL pointer dereference at virtual address 00000038
>  Internal error: Oops: 96000004 [#1] PREEMPT SMP
>  Modules linked in: [last unloaded: arm_spe_pmu]
>  PC is at ctx_resched+0x38/0xe8
>  LR is at perf_event_exec+0x20c/0x278
>  [...]
>  ctx_resched+0x38/0xe8
>  perf_event_exec+0x20c/0x278
>  setup_new_exec+0x88/0x118
>  load_elf_binary+0x26c/0x109c
>  search_binary_handler+0x90/0x298
>  do_execveat_common.isra.14+0x540/0x618
>  SyS_execve+0x38/0x48
> 
> since the software context has been freed and the ctx.pmu->pmu_disable_count
> field has been set to NULL.
> 
> This patch fixes the problem by avoiding the freeing of static PMU contexts
> altogether. Whilst the sharing of dynamic contexts is questionable, this
> actually requires the caller to share their context pointer explicitly
> and so the burden is on them to manage the object lifetime.
> 
> Cc: Peter Zijlstra <peterz at infradead.org>
> Fixes: 1fd7e4169954 ("perf/core: Remove perf_cpu_context::unique_pmu")
> Reported-by: Kim Phillips <kim.phillips at arm.com>
> Signed-off-by: Will Deacon <will.deacon at arm.com>

FWIW:

Acked-by: Mark Rutland <mark.rutland at arm.com>

Mark.

> ---
>  kernel/events/core.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index 6bc21e202ae4..243bfc68d0fe 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -8955,6 +8955,14 @@ static struct perf_cpu_context __percpu *find_pmu_context(int ctxn)
>  
>  static void free_pmu_context(struct pmu *pmu)
>  {
> +	/*
> +	 * Static contexts such as perf_sw_context have a global lifetime
> +	 * and may be shared between different PMUs. Avoid freeing them
> +	 * when a single PMU is going away.
> +	 */
> +	if (pmu->task_ctx_nr > perf_invalid_context)
> +		return;
> +
>  	mutex_lock(&pmus_lock);
>  	free_percpu(pmu->pmu_cpu_context);
>  	mutex_unlock(&pmus_lock);
> -- 
> 2.1.4
>