Re: [PATCH] x86/hw_breakpoints: Return -EINVAL in default case in hw_breakpoint_arch_parse

[Nathan Chancellor]

On Thu, Mar 07, 2019 at 02:27:56PM -0700, Nathan Chancellor wrote:
> When building with -Wsometimes-uninitialized, Clang warns:
> 
> arch/x86/kernel/hw_breakpoint.c:355:2: warning: variable 'align' is used
> uninitialized whenever switch default is taken
> [-Wsometimes-uninitialized]
> 
> It's not wrong but we can see that arch_build_bp_info will only ever
> initialize hw->len to one of the specified switch cases. We can easily
> silence Clang by just returning -EINVAL in the default case so that we
> can never use align without initializing it first.
> 
> Link: https://github.com/ClangBuiltLinux/linux/issues/392
> Suggested-by: Nick Desaulniers <ndesaulniers@xxxxxxxxxx>
> Signed-off-by: Nathan Chancellor <natechancellor@xxxxxxxxx>
> ---
>  arch/x86/kernel/hw_breakpoint.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c
> index ff9bfd40429e..d73083021002 100644
> --- a/arch/x86/kernel/hw_breakpoint.c
> +++ b/arch/x86/kernel/hw_breakpoint.c
> @@ -354,6 +354,7 @@ int hw_breakpoint_arch_parse(struct perf_event *bp,
>  #endif
>  	default:
>  		WARN_ON_ONCE(1);
> +		return -EINVAL;
>  	}
>  
>  	/*
> -- 
> 2.21.0
> 

Gentle ping (if there was a response to this, I didn't receive it). I
know I sent it in the middle of a merge window so I get if it slipped
through the cracks.

Thanks,
Nathan