Re: [PATCH v1 1/2] perf tools: Improve handling of hybrid PMUs in perf_event_attr__fprintf
From: James Clark
Date: Wed Mar 05 2025 - 06:06:47 EST
On 05/03/2025 8:37 am, Ian Rogers wrote:
Support the PMU name from the legacy hardware and hw_cache PMU
extended types. Remove some macros and make variables more intention
revealing, rather than just being called "value".
Before:
```
$ perf stat -vv -e instructions true
...
------------------------------------------------------------
perf_event_attr:
type 0 (PERF_TYPE_HARDWARE)
size 136
config 0xa00000001
sample_type IDENTIFIER
read_format TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING
disabled 1
inherit 1
enable_on_exec 1
exclude_guest 1
------------------------------------------------------------
sys_perf_event_open: pid 181636 cpu -1 group_fd -1 flags 0x8 = 5
------------------------------------------------------------
perf_event_attr:
type 0 (PERF_TYPE_HARDWARE)
size 136
config 0x400000001
sample_type IDENTIFIER
read_format TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING
disabled 1
inherit 1
enable_on_exec 1
exclude_guest 1
------------------------------------------------------------
sys_perf_event_open: pid 181636 cpu -1 group_fd -1 flags 0x8 = 6
...
```
After:
```
$ perf stat -vv -e instructions true
...
------------------------------------------------------------
perf_event_attr:
type 0 (PERF_TYPE_HARDWARE)
size 136
config 0xa00000001 (cpu_atom/PERF_COUNT_HW_INSTRUCTIONS/)
Seems like a good idea, I'm always decoding these by eye.
Reviewed-by: James Clark <james.clark@xxxxxxxxxx>