From: Jani Nikula <jani.nikula@intel.com>
To: Pranay Samala <pranay.samala@intel.com>, igt-dev@lists.freedesktop.org
Cc: karthik.b.s@intel.com, kunal1.joshi@intel.com,
sameer.lattannavar@intel.com, pranay.samala@intel.com,
Leo Li <sunpeng.li@amd.com>, Uma Shankar <uma.shankar@intel.com>,
Ramanaidu Naladala <ramanaidu.naladala@intel.com>
Subject: Re: [PATCH i-g-t 1/2] lib/igt_sysfs: Update DRM debug severity handling to use bitmask for verbosity control
Date: Tue, 11 Mar 2025 10:49:14 +0200 [thread overview]
Message-ID: <87ikoggdf9.fsf@intel.com> (raw)
In-Reply-To: <20250310171539.273699-2-pranay.samala@intel.com>
On Mon, 10 Mar 2025, Pranay Samala <pranay.samala@intel.com> wrote:
> This approach ensures precise control by modifying
> specific bits in the DRM debug mask.
>
> Updated the DRM debug severity logic to use bitwise
> operations to unset unwanted bits in the debug mask.
> The function now only sets the bits that are passed,
> allowing for more granular control over the logging output.
Please don't call it "severity" when drm debug is about "category".
You can't say one category is more "severe" than another.
>
> Cc: Leo Li <sunpeng.li@amd.com>
> Cc: Jani Nikula <jani.nikula@intel.com>
> Cc: Uma Shankar <uma.shankar@intel.com>
> Cc: Kunal Joshi <kunal1.joshi@intel.com>
> Cc: Karthik B S <karthik.b.s@intel.com>
> Cc: Ramanaidu Naladala <ramanaidu.naladala@intel.com>
> Cc: Sameer Lattannavar <sameer.lattannavar@intel.com>
>
> Fixes: 56b91193b825 ("lib/igt_sysfs: Implement dynamic adjustment of debug log level")
> Signed-off-by: Pranay Samala <pranay.samala@intel.com>
> ---
> lib/igt_sysfs.c | 54 ++++++++++++++++++++++++++-----------------------
> lib/igt_sysfs.h | 19 ++++++++++++++---
> 2 files changed, 45 insertions(+), 28 deletions(-)
>
> diff --git a/lib/igt_sysfs.c b/lib/igt_sysfs.c
> index 2e4c2ee63..2a3cb68e4 100644
> --- a/lib/igt_sysfs.c
> +++ b/lib/igt_sysfs.c
> @@ -443,69 +443,71 @@ int igt_sysfs_drm_module_params_open(void)
> return open(path, O_RDONLY);
> }
>
> -static int log_level = -1;
> +static int log_severity = -1;
>
> /**
> - * igt_drm_debug_level_get:
> + * igt_drm_debug_severity_get:
> *
> - * This reads the current debug log level of the machine on
> + * This reads the current debug log severity of the machine on
> * which the test is currently executing.
> *
> * Returns:
> - * The current log level, or -1 on error.
> + * The current log severity, or -1 on error.
> */
> -int igt_drm_debug_level_get(int dir)
> +int igt_drm_debug_severity_get(int dir)
> {
> char buf[20];
>
> - if (log_level >= 0)
> - return log_level;
> + if (log_severity >= 0)
> + return log_severity;
>
> - if (igt_sysfs_read(dir, "debug", buf, sizeof(buf) - 1) < 0)
> + if (igt_sysfs_read(dir, "debug", buf, sizeof(buf)) < 0)
> return -1;
>
> return atoi(buf);
> }
>
> /**
> - * igt_drm_debug_level_reset:
> + * igt_drm_debug_severity_reset:
> *
> - * This modifies the current debug log level of the machine
> + * This modifies the current debug log severity of the machine
> * to the default value post-test.
> *
> */
> -void igt_drm_debug_level_reset(void)
> +void igt_drm_debug_severity_reset(void)
> {
> char buf[20];
> int dir;
>
> - if (log_level < 0)
> + if (log_severity < 0)
> return;
>
> dir = igt_sysfs_drm_module_params_open();
> if (dir < 0)
> return;
>
> - igt_debug("Resetting DRM debug level to %d\n", log_level);
> - snprintf(buf, sizeof(buf), "%d", log_level);
> + igt_debug("Resetting DRM debug severity to 0x%x\n", log_severity);
> + snprintf(buf, sizeof(buf), "%d", log_severity);
> igt_assert(igt_sysfs_set(dir, "debug", buf));
>
> close(dir);
> }
>
> -static void igt_drm_debug_level_reset_exit_handler(int sig)
> +static void igt_drm_debug_severity_reset_exit_handler(int sig)
> {
> - igt_drm_debug_level_reset();
> + igt_drm_debug_severity_reset();
> }
>
> /**
> - * igt_drm_debug_level_update:
> - * @debug_level: new debug level to set
> + * igt_drm_debug_severity_update:
> + * @debug_severity: new debug severity to set
> *
> - * This modifies the current drm debug log level to the new value.
> + * This modifies the current drm debug log severity to the new value.
> */
> -void igt_drm_debug_level_update(unsigned int new_log_level)
> +
> +void igt_drm_debug_severity_update(enum drm_debug_category category_to_set)
This should accept a mask instead of a single category, and the function
and parameter naming should reflect that.
> {
> + unsigned int new_log_severity;
> char buf[20];
> int dir;
>
> @@ -513,14 +515,16 @@ void igt_drm_debug_level_update(unsigned int new_log_level)
> if (dir < 0)
> return;
>
> - log_level = igt_drm_debug_level_get(dir);
> - if (log_level < 0) {
> + log_severity = igt_drm_debug_severity_get(dir);
> + if (log_severity < 0) {
> close(dir);
> return;
> }
This doesn't seem to handle nested updates. The reset will put this back
to the previous value, not the original value.
>
> - igt_debug("Setting DRM debug level to %d\n", new_log_level);
> - snprintf(buf, sizeof(buf), "%d", new_log_level);
> + new_log_severity = category_to_set;
> +
> + igt_debug("Setting DRM debug severity to 0x%x\n", new_log_severity);
> + snprintf(buf, sizeof(buf), "%d", new_log_severity);
> igt_assert(igt_sysfs_set(dir, "debug", buf));
>
> close(dir);
> @@ -529,7 +533,7 @@ void igt_drm_debug_level_update(unsigned int new_log_level)
> * TODO: Check whether multiple exit handlers will get installed,
> * if we call this api multiple times
> */
> - igt_install_exit_handler(igt_drm_debug_level_reset_exit_handler);
> + igt_install_exit_handler(igt_drm_debug_severity_reset_exit_handler);
> }
>
> /**
> diff --git a/lib/igt_sysfs.h b/lib/igt_sysfs.h
> index 86345f3d1..c2e15320d 100644
> --- a/lib/igt_sysfs.h
> +++ b/lib/igt_sysfs.h
> @@ -143,9 +143,22 @@ void igt_sysfs_set_boolean(int dir, const char *attr, bool value);
> void bind_fbcon(bool enable);
> void fbcon_blink_enable(bool enable);
>
> -void igt_drm_debug_level_update(unsigned int new_log_level);
> -void igt_drm_debug_level_reset(void);
> -int igt_drm_debug_level_get(int dir);
> +enum drm_debug_category {
> + DRM_UT_CORE = 1 << 0,
> + DRM_UT_DRIVER = 1 << 1,
> + DRM_UT_KMS = 1 << 2,
> + DRM_UT_PRIME = 1 << 3,
> + DRM_UT_ATOMIC = 1 << 4,
> + DRM_UT_VBL = 1 << 5,
> + DRM_UT_STATE = 1 << 6,
> + DRM_UT_LEASE = 1 << 7,
> + DRM_UT_DP = 1 << 8,
> + DRM_UT_DRMRES = 1 << 9,
> +};
> +
> +void igt_drm_debug_severity_update(enum drm_debug_category category_to_set);
> +void igt_drm_debug_severity_reset(void);
> +int igt_drm_debug_severity_get(int dir);
> int igt_sysfs_drm_module_params_open(void);
>
> /**
--
Jani Nikula, Intel
next prev parent reply other threads:[~2025-03-11 8:49 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-10 17:15 [PATCH i-g-t 0/2] Refactor DRM Debug Severity Handling for enhanced precision Pranay Samala
2025-03-10 17:15 ` [PATCH i-g-t 1/2] lib/igt_sysfs: Update DRM debug severity handling to use bitmask for verbosity control Pranay Samala
2025-03-11 8:49 ` Jani Nikula [this message]
2025-03-10 17:15 ` [PATCH i-g-t 2/2] tests/kms: Simplify DRM debug severity update Pranay Samala
2025-03-11 8:55 ` Jani Nikula
2025-03-11 0:47 ` ✓ Xe.CI.BAT: success for Refactor DRM Debug Severity Handling for enhanced precision Patchwork
2025-03-11 1:05 ` ✓ i915.CI.BAT: " Patchwork
2025-03-11 20:44 ` ✗ Xe.CI.Full: 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=87ikoggdf9.fsf@intel.com \
--to=jani.nikula@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=karthik.b.s@intel.com \
--cc=kunal1.joshi@intel.com \
--cc=pranay.samala@intel.com \
--cc=ramanaidu.naladala@intel.com \
--cc=sameer.lattannavar@intel.com \
--cc=sunpeng.li@amd.com \
--cc=uma.shankar@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.