Re: [Patch v10 09/10] perf tests: Add x86 eGPRs/SSP registers sampling test

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The newly added validation function `validate_gp_reg_sampling` executes a single-shot pipeline for `perf record`, bypassing the established `perf_record_with_retry` wrapper. This makes the test susceptible to intermittent failures (flakiness) in CI environments and prevents it from utilizing the fast-path optimization.
--

--- Patch [9]: [Patch v10 09/10] perf tests: Add x86 eGPRs/SSP registers sampling test ---
commit a4a8e0ee768722ad01317e4cd04c08a71c39eecf
Author: Dapeng Mi <dapeng1.mi@linux.intel.com>

perf tests: Add x86 eGPRs/SSP registers sampling test

Validate that eGPRs (extended general purpose registers) and SSP
(shadow stack pointer) registers can be sampled correctly on x86
platforms with hardware support.

> diff --git a/tools/perf/tests/shell/record.sh b/tools/perf/tests/shell/record.sh
> index 625240809fce3..3c17f68ceb929 100755
> --- a/tools/perf/tests/shell/record.sh
> +++ b/tools/perf/tests/shell/record.sh

[ ... ]

> +validate_gp_reg_sampling() {
> +  local regs_opt="$1"
> +  local regs_value="$2"
> +  local script_field="$3"
> +  local sample_output
> +  shift 3
> +
> +  if ! sample_output=$(perf record -o - "${regs_opt}=${regs_value}" \
> +    -e br_inst_retired.near_call -c 1000 --per-thread ${testprog} 2> /dev/null \
> +    | perf script -F ip,sym,"${script_field}" -i - 2> /dev/null)

[Severity: Medium]
Will this single-shot pipeline cause intermittent failures in CI environments
under heavy load?

Other tests in this file appear to use the perf_record_with_retry wrapper 
to mitigate flakiness. That wrapper provides a fast-path optimization and
scales up the recording duration if the system load prevents sufficient
hardware events from being generated immediately.

Since validate_gp_reg_sampling() executes perf record directly without a retry
loop, could it fail spuriously if the workload doesn't generate enough events
in the initial window?

> +  then
> +    return 1
> +  fi

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260721072707.3787848-1-dapeng1.mi@linux.intel.com?part=9