[PATCH] perf/tool: Add remaining branch filters to perf record

[PATCH] perf/tool: Add remaining branch filters to perf record

[Anshuman Khandual]

This adds all remaining branch filters i.e no_cycles, no_flags and hw_index
to perf record tool. While here, also updates the documentation.

Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: linux-perf-users@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
---
 tools/perf/Documentation/perf-record.txt | 5 +++++
 tools/perf/util/parse-branch-options.c   | 3 +++
 2 files changed, 8 insertions(+)

diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index e41ae950fdc3..7803c0c4c8f9 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -388,6 +388,7 @@ following filters are defined:
         - any_call: any function call or system call
         - any_ret: any function return or system call return
         - ind_call: any indirect branch
+        - ind_jmp: any indirect jump
         - call: direct calls, including far (to/from kernel) calls
         - u:  only when the branch target is at the user level
         - k: only when the branch target is in the kernel
@@ -396,6 +397,10 @@ following filters are defined:
 	- no_tx: only when the target is not in a hardware transaction
 	- abort_tx: only when the target is a hardware transaction abort
 	- cond: conditional branches
+	- call_stack: save call stack
+	- no_flags: don't save branch flags e.g prediction, misprediction etc
+	- no_cycles: don't save branch cycles
+	- hw_index: save branch hardware index
 	- save_type: save branch type during sampling in case binary is not available later
 		     For the platforms with Intel Arch LBR support (12th-Gen+ client or
 		     4th-Gen Xeon+ server), the save branch type is unconditionally enabled
diff --git a/tools/perf/util/parse-branch-options.c b/tools/perf/util/parse-branch-options.c
index 31faf2bb49ff..fd67d204d720 100644
--- a/tools/perf/util/parse-branch-options.c
+++ b/tools/perf/util/parse-branch-options.c
@@ -30,8 +30,11 @@ static const struct branch_mode branch_modes[] = {
 	BRANCH_OPT("cond", PERF_SAMPLE_BRANCH_COND),
 	BRANCH_OPT("ind_jmp", PERF_SAMPLE_BRANCH_IND_JUMP),
 	BRANCH_OPT("call", PERF_SAMPLE_BRANCH_CALL),
+	BRANCH_OPT("no_flags", PERF_SAMPLE_BRANCH_NO_FLAGS),
+	BRANCH_OPT("no_cycles", PERF_SAMPLE_BRANCH_NO_CYCLES),
 	BRANCH_OPT("save_type", PERF_SAMPLE_BRANCH_TYPE_SAVE),
 	BRANCH_OPT("stack", PERF_SAMPLE_BRANCH_CALL_STACK),
+	BRANCH_OPT("hw_index", PERF_SAMPLE_BRANCH_HW_INDEX),
 	BRANCH_OPT("priv", PERF_SAMPLE_BRANCH_PRIV_SAVE),
 	BRANCH_END
 };
-- 
2.25.1

Re: [PATCH] perf/tool: Add remaining branch filters to perf record

[Arnaldo Carvalho de Melo]

Em Mon, Dec 05, 2022 at 12:14:43PM +0530, Anshuman Khandual escreveu:
> This adds all remaining branch filters i.e no_cycles, no_flags and hw_index
> to perf record tool. While here, also updates the documentation.

Thanks, applied.

- Arnaldo

 
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
> Cc: linux-perf-users@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org
> Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
> ---
>  tools/perf/Documentation/perf-record.txt | 5 +++++
>  tools/perf/util/parse-branch-options.c   | 3 +++
>  2 files changed, 8 insertions(+)
> 
> diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
> index e41ae950fdc3..7803c0c4c8f9 100644
> --- a/tools/perf/Documentation/perf-record.txt
> +++ b/tools/perf/Documentation/perf-record.txt
> @@ -388,6 +388,7 @@ following filters are defined:
>          - any_call: any function call or system call
>          - any_ret: any function return or system call return
>          - ind_call: any indirect branch
> +        - ind_jmp: any indirect jump
>          - call: direct calls, including far (to/from kernel) calls
>          - u:  only when the branch target is at the user level
>          - k: only when the branch target is in the kernel
> @@ -396,6 +397,10 @@ following filters are defined:
>  	- no_tx: only when the target is not in a hardware transaction
>  	- abort_tx: only when the target is a hardware transaction abort
>  	- cond: conditional branches
> +	- call_stack: save call stack
> +	- no_flags: don't save branch flags e.g prediction, misprediction etc
> +	- no_cycles: don't save branch cycles
> +	- hw_index: save branch hardware index
>  	- save_type: save branch type during sampling in case binary is not available later
>  		     For the platforms with Intel Arch LBR support (12th-Gen+ client or
>  		     4th-Gen Xeon+ server), the save branch type is unconditionally enabled
> diff --git a/tools/perf/util/parse-branch-options.c b/tools/perf/util/parse-branch-options.c
> index 31faf2bb49ff..fd67d204d720 100644
> --- a/tools/perf/util/parse-branch-options.c
> +++ b/tools/perf/util/parse-branch-options.c
> @@ -30,8 +30,11 @@ static const struct branch_mode branch_modes[] = {
>  	BRANCH_OPT("cond", PERF_SAMPLE_BRANCH_COND),
>  	BRANCH_OPT("ind_jmp", PERF_SAMPLE_BRANCH_IND_JUMP),
>  	BRANCH_OPT("call", PERF_SAMPLE_BRANCH_CALL),
> +	BRANCH_OPT("no_flags", PERF_SAMPLE_BRANCH_NO_FLAGS),
> +	BRANCH_OPT("no_cycles", PERF_SAMPLE_BRANCH_NO_CYCLES),
>  	BRANCH_OPT("save_type", PERF_SAMPLE_BRANCH_TYPE_SAVE),
>  	BRANCH_OPT("stack", PERF_SAMPLE_BRANCH_CALL_STACK),
> +	BRANCH_OPT("hw_index", PERF_SAMPLE_BRANCH_HW_INDEX),
>  	BRANCH_OPT("priv", PERF_SAMPLE_BRANCH_PRIV_SAVE),
>  	BRANCH_END
>  };
> -- 
> 2.25.1

-- 

- Arnaldo