Re: [PATCH V2 4/8] perf: Capture branch privilege information

[James Clark]

On 09/03/2022 03:36, Anshuman Khandual wrote:
> Platforms like arm64 could capture privilege level information for all the
> branch records. Hence this adds a new element in the struct branch_entry to
> record the privilege level information, which could be requested through a
> new event.attr.branch_sample_type based flag PERF_SAMPLE_BRANCH_PRIV_SAVE.
> This flag helps user choose whether privilege information is captured.
> 
> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> Cc: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
> Cc: Mark Rutland <mark.rutland@xxxxxxx>
> Cc: Alexander Shishkin <alexander.shishkin@xxxxxxxxxxxxxxx>
> Cc: Jiri Olsa <jolsa@xxxxxxxxxx>
> Cc: Namhyung Kim <namhyung@xxxxxxxxxx>
> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> Cc: Will Deacon <will@xxxxxxxxxx>
> Cc: linux-arm-kernel@xxxxxxxxxxxxxxxxxxx
> Cc: linux-perf-users@xxxxxxxxxxxxxxx
> Cc: linux-kernel@xxxxxxxxxxxxxxx
> Signed-off-by: Anshuman Khandual <anshuman.khandual@xxxxxxx>
> ---
>  include/uapi/linux/perf_event.h | 13 ++++++++++++-
>  1 file changed, 12 insertions(+), 1 deletion(-)
> 
> diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
> index d29280adc3c4..0e96e2017f68 100644
> --- a/include/uapi/linux/perf_event.h
> +++ b/include/uapi/linux/perf_event.h
> @@ -204,6 +204,8 @@ enum perf_branch_sample_type_shift {
>  
>  	PERF_SAMPLE_BRANCH_HW_INDEX_SHIFT	= 17, /* save low level index of raw branch records */
>  
> +	PERF_SAMPLE_BRANCH_PRIV_SAVE_SHIFT	= 18, /* save privillege mode */
> +
>  	PERF_SAMPLE_BRANCH_MAX_SHIFT		/* non-ABI */
>  };
>  
> @@ -233,6 +235,8 @@ enum perf_branch_sample_type {
>  
>  	PERF_SAMPLE_BRANCH_HW_INDEX	= 1U << PERF_SAMPLE_BRANCH_HW_INDEX_SHIFT,
>  
> +	PERF_SAMPLE_BRANCH_PRIV_SAVE	= 1U << PERF_SAMPLE_BRANCH_PRIV_SAVE_SHIFT,
> +
>  	PERF_SAMPLE_BRANCH_MAX		= 1U << PERF_SAMPLE_BRANCH_MAX_SHIFT,
>  };
>  
> @@ -271,6 +275,12 @@ enum {
>  	PERF_BR_NEW_MAX,
>  };
>  
> +enum {
> +	PERF_BR_USER	= 0,
> +	PERF_BR_KERNEL	= 1,
> +	PERF_BR_HV	= 2,
> +};

0 should be "PERF_BR_PRIV_UNKNOWN" so userspace knows if it was not enabled
otherwise it will look like all samples are PERF_BR_USER when actually
priv type recording was just disabled.

I think it's not even always possible to go backwards from a sample to
work out what the event attributes were so this can be interpreted (taking
all of perf script and every corner case into account).

Starting at 0=UNKNOWN is consistent with the other fields and makes parsing
it a whole lot easier.

James

> +
>  #define PERF_SAMPLE_BRANCH_PLM_ALL \
>  	(PERF_SAMPLE_BRANCH_USER|\
>  	 PERF_SAMPLE_BRANCH_KERNEL|\
> @@ -1386,7 +1396,8 @@ struct perf_branch_entry {
>  		cycles:16,  /* cycle count to last branch */
>  		type:4,     /* branch type */
>  		new_type:4, /* additional branch type */
> -		reserved:36;
> +		priv:2,     /* privilege level */
> +		reserved:34;
>  };
>  
>  union perf_sample_weight {