[igt-dev] [PATCH i-g-t v3] tests/i915/i915_pm_dc: Add DC9 igt test
Petri Latvala
petri.latvala at intel.com
Mon Feb 8 06:48:02 UTC 2021
On Fri, Feb 05, 2021 at 02:45:22PM +0530, Jigar Bhatt wrote:
> DC9 igt test validation depends on DC9 Ready and DC{5,6} counters reset.
> When Display Engine enters to DC9, it resets DMC f/w and DC5/DC6 counter.
> We don't have a DC9 counter unlike DC{5,6} therefore this tests uses "DC9 Ready"
> from i915_pm_req_res_info debugfs and DC{5,6} counter reset method to validate DC9.
>
> Signed-off-by: Jigar Bhatt <jigar.bhatt at intel.com>
> ---
> tests/i915/i915_pm_dc.c | 25 +++++++++++++++++++++++++
> 1 file changed, 25 insertions(+)
>
> diff --git a/tests/i915/i915_pm_dc.c b/tests/i915/i915_pm_dc.c
> index 30c6024..957d04a 100644
> --- a/tests/i915/i915_pm_dc.c
> +++ b/tests/i915/i915_pm_dc.c
> @@ -384,6 +384,27 @@ static void test_dc_state_dpms(data_t *data, int dc_flag)
> cleanup_dc_dpms(data);
> }
>
> +static bool check_dc9(uint32_t debugfs_fd, int prev_dc5, int prev_dc6)
> +{
> + char buf[4096];
> + igt_skip_on_f(!igt_debugfs_simple_read(debugfs_fd, "i915_pm_req_res_info", buf, sizeof(buf)),"skipping DC9 test");
> + igt_assert_f(strstr(buf, "DC9 Ready : yes"),"DC9 is not ready");
These messages need to end in newline
> + /*
> + * since we do not have DC9 Counter,
> + * so we rely on dc5/dc6 counter reset to check if Display Engine was in DC9.
> + */
> + return igt_wait((read_dc_counter(debugfs_fd, CHECK_DC5)<prev_dc5 || read_dc_counter(debugfs_fd, CHECK_DC6)<prev_dc6),3000,100);
> +}
> +static void test_dc9_dpms(data_t *data)
> +{
> + require_dc_counter(data->debugfs_fd, CHECK_DC5);
> + require_dc_counter(data->debugfs_fd, CHECK_DC6);
> + int dc5_prev=read_dc_counter(data->debugfs_fd, CHECK_DC5);
> + int dc6_prev=read_dc_counter(data->debugfs_fd, CHECK_DC6);
> + dpms_off(data);
> + igt_assert_f(check_dc9(data->debugfs_fd,dc5_prev,dc6_prev),"\n Not in DC9");
Same here. And why does it begin with a newline?
--
Petri Latvala
More information about the igt-dev
mailing list