Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Mauro Carvalho Chehab <mauro.chehab@linux.intel.com>
To: Kamil Konieczny <kamil.konieczny@linux.intel.com>
Cc: igt-dev@lists.freedesktop.org
Subject: Re: [igt-dev] [PATCH i-g-t 1/2] runner/settings: add dump-gpu-on-timeout option
Date: Tue, 19 Sep 2023 09:20:30 +0200	[thread overview]
Message-ID: <20230919092030.3b2bcec1@maurocar-mobl2> (raw)
In-Reply-To: <20230714153946.36448-2-kamil.konieczny@linux.intel.com>

On Fri, 14 Jul 2023 17:39:45 +0200
Kamil Konieczny <kamil.konieczny@linux.intel.com> wrote:

> Add new option --dump-gpu-on-timeout for dumping error state
> of GPU in case of inactivity or per-test timeout occurs. This
> will help in faster problem diagnose from CI runs.
> 
> Signed-off-by: Kamil Konieczny <kamil.konieczny@linux.intel.com>

The change itself LGTM. I'm wondering if it makes sense in this case
to have an unit test to check it.

For the patch:

Reviewed-by: Mauro Carvalho Chehab <mchehab@kernel.org>

> ---
>  runner/settings.c | 12 +++++++++++-
>  runner/settings.h |  1 +
>  2 files changed, 12 insertions(+), 1 deletion(-)
> 
> diff --git a/runner/settings.c b/runner/settings.c
> index 23aa82963..880aefbd6 100644
> --- a/runner/settings.c
> +++ b/runner/settings.c
> @@ -28,6 +28,7 @@ enum {
>  	OPT_CODE_COV_SCRIPT,
>  	OPT_ENABLE_CODE_COVERAGE,
>  	OPT_COV_RESULTS_PER_TEST,
> +	OPT_DUMP_GPU_ON_TIMEOUT,
>  	OPT_VERSION,
>  	OPT_PRUNE_MODE,
>  	OPT_HELP = 'h',
> @@ -297,6 +298,10 @@ static const char *usage_str =
>  	"                        Requires --collect-script FILENAME\n"
>  	"  --collect-script FILENAME\n"
>  	"                        Use FILENAME as script to collect code coverage data.\n"
> +	"  --dump-gpu-on-timeout FILENAME\n"
> +	"                        Use /sys/class/drm/card*/error (default) or (if default is empty)\n"
> +	"                        /sys/kernel/debug/dri/*/FILENAME for reading GPU error state\n"
> +	"                        in case of inactivity/per-test timeout occurs.\n"
>  	"\n"
>  	"  [test_root]           Directory that contains the IGT tests. The environment\n"
>  	"                        variable IGT_TEST_ROOT will be used if set, overriding\n"
> @@ -654,6 +659,7 @@ bool parse_options(int argc, char **argv,
>  		{"collect-code-cov", no_argument, NULL, OPT_ENABLE_CODE_COVERAGE},
>  		{"coverage-per-test", no_argument, NULL, OPT_COV_RESULTS_PER_TEST},
>  		{"collect-script", required_argument, NULL, OPT_CODE_COV_SCRIPT},
> +		{"dump-gpu-on-timeout", required_argument, NULL, OPT_DUMP_GPU_ON_TIMEOUT},
>  		{"multiple-mode", no_argument, NULL, OPT_MULTIPLE},
>  		{"inactivity-timeout", required_argument, NULL, OPT_TIMEOUT},
>  		{"per-test-timeout", required_argument, NULL, OPT_PER_TEST_TIMEOUT},
> @@ -740,7 +746,9 @@ bool parse_options(int argc, char **argv,
>  		case OPT_CODE_COV_SCRIPT:
>  			settings->code_coverage_script = bin_path(optarg);
>  			break;
> -
> +		case OPT_DUMP_GPU_ON_TIMEOUT:
> +			settings->dump_gpu_on_timeout = bin_path(optarg);
> +			break;
>  		case OPT_MULTIPLE:
>  			settings->multiple_mode = true;
>  			break;
> @@ -1039,6 +1047,7 @@ bool serialize_settings(struct settings *settings)
>  	SERIALIZE_LINE(f, settings, enable_code_coverage, "%d");
>  	SERIALIZE_LINE(f, settings, cov_results_per_test, "%d");
>  	SERIALIZE_LINE(f, settings, code_coverage_script, "%s");
> +	SERIALIZE_LINE(f, settings, dump_gpu_on_timeout, "%s");
>  
>  	if (settings->sync) {
>  		fflush(f);
> @@ -1102,6 +1111,7 @@ bool read_settings_from_file(struct settings *settings, FILE *f)
>  		PARSE_LINE(settings, name, val, enable_code_coverage, numval);
>  		PARSE_LINE(settings, name, val, cov_results_per_test, numval);
>  		PARSE_LINE(settings, name, val, code_coverage_script, val ? strdup(val) : NULL);
> +		PARSE_LINE(settings, name, val, dump_gpu_on_timeout, val ? strdup(val) : NULL);
>  
>  		printf("Warning: Unknown field in settings file: %s = %s\n",
>  		       name, val);
> diff --git a/runner/settings.h b/runner/settings.h
> index 819c34602..ab9e4c630 100644
> --- a/runner/settings.h
> +++ b/runner/settings.h
> @@ -72,6 +72,7 @@ struct settings {
>  	char *code_coverage_script;
>  	bool enable_code_coverage;
>  	bool cov_results_per_test;
> +	char *dump_gpu_on_timeout;
>  };
>  
>  /**

  reply	other threads:[~2023-09-19  7:20 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-14 15:39 [igt-dev] [PATCH i-g-t 0/2] RFC: runner: dump GPU error state on timeout Kamil Konieczny
2023-07-14 15:39 ` [igt-dev] [PATCH i-g-t 1/2] runner/settings: add dump-gpu-on-timeout option Kamil Konieczny
2023-09-19  7:20   ` Mauro Carvalho Chehab [this message]
2023-07-14 15:39 ` [igt-dev] [PATCH i-g-t 2/2] runner/executor: write GPU error on timeout Kamil Konieczny
2023-09-19  7:29   ` Mauro Carvalho Chehab
2023-09-19  8:09     ` Kamil Konieczny
2023-07-14 15:58 ` [igt-dev] ✗ GitLab.Pipeline: warning for RFC: runner: dump GPU error state " Patchwork
2023-07-14 16:29 ` [igt-dev] ✓ Fi.CI.BAT: success " Patchwork
2023-07-14 17:01 ` [igt-dev] ○ CI.xeBAT: info " Patchwork
2023-07-14 18:30 ` [igt-dev] ✗ Fi.CI.IGT: failure " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230919092030.3b2bcec1@maurocar-mobl2 \
    --to=mauro.chehab@linux.intel.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=kamil.konieczny@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox