From: Pranay Samala <pranay.samala@intel.com>
To: igt-dev@lists.freedesktop.org
Cc: karthik.b.s@intel.com, sameer.lattannavar@intel.com,
pranay.samala@intel.com
Subject: [PATCH i-g-t 5/7] lib/igt_pm: Add power/wakeup_active_count accessor
Date: Mon, 7 Sep 2026 19:52:57 +0530 [thread overview]
Message-ID: <20260907142259.750528-6-pranay.samala@intel.com> (raw)
In-Reply-To: <20260907142259.750528-1-pranay.samala@intel.com>
Add igt_pm_get_wakeup_active_count(), which reads how many times the
device's wakeup source has been activated.
wakeup_source_activate() bumps this unconditionally, so it counts runtime
resumes too, on both the PCIe (pci_pme_wakeup()) and ACPI GPE
(pci_acpi_wake_dev()) delivery routes. An increment across a suspended
window therefore means the device signalled the resume itself, i.e. a PME
arrived, rather than the host waking it.
Neither alternative works: power/wakeup_count only moves under
events_check_enabled, which the PM core sets for system suspend and
hibernate, and PME_Status is cleared by the PCI/PM core as it handles the
wakeup, so it cannot be sampled afterwards.
Parsing uses a checked sscanf() rather than an asserted one, since the
counter reads back as a bare newline when the device has no wakeup source,
that is when power/wakeup is disabled. That has to give 0, not a failure.
Assisted-by: GitHub_Copilot:claude-opus-5
Signed-off-by: Pranay Samala <pranay.samala@intel.com>
---
lib/igt_pm.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++++
lib/igt_pm.h | 1 +
2 files changed, 49 insertions(+)
diff --git a/lib/igt_pm.c b/lib/igt_pm.c
index a3ad80580..a9fd9610a 100644
--- a/lib/igt_pm.c
+++ b/lib/igt_pm.c
@@ -1498,6 +1498,34 @@ int igt_pm_get_runtime_usage(struct pci_device *pci_dev)
return usage;
}
+/*
+ * Read an unsigned power attribute of @pci_dev, returning 0 if the attribute
+ * does not exist or does not hold a number.
+ *
+ * The latter is not hypothetical: the wakeup counters exist for every wakeup
+ * capable device, but the PM core emits a bare newline for them while the
+ * device has no wakeup source, i.e. while power/wakeup is disabled.
+ */
+static uint64_t igt_pm_read_power_attr_u64(struct pci_device *pci_dev,
+ const char *attr)
+{
+ char buf[64];
+ uint64_t val;
+ int fd;
+
+ fd = __igt_pm_get_power_attr_fd(pci_dev, attr, O_RDONLY);
+ if (fd < 0)
+ return 0;
+
+ if (!igt_pm_read_power_attr(fd, buf, sizeof(buf), false) ||
+ sscanf(buf, "%" SCNu64, &val) != 1)
+ val = 0;
+
+ close(fd);
+
+ return val;
+}
+
#define IGT_PM_WAKEUP_ENABLED_STR "enabled\n"
#define IGT_PM_WAKEUP_DISABLED_STR "disabled\n"
@@ -1631,6 +1659,26 @@ void igt_pm_restore_wakeup(void)
__igt_pm_wakeup.pci_dev = NULL;
}
+/**
+ * igt_pm_get_wakeup_active_count:
+ * @pci_dev: PCI device struct
+ *
+ * Reads power/wakeup_active_count, the number of times the device's wakeup
+ * source was activated. An increment across a runtime suspended window means
+ * the resume was signalled by the device, i.e. a PME arrived, rather than
+ * initiated by something host side such as HPD polling.
+ *
+ * Use this and not power/wakeup_count, which is only incremented under
+ * events_check_enabled and so never moves for a runtime resume.
+ *
+ * Return: the wakeup active count, or 0 if the device is not wakeup capable or
+ * has power/wakeup disabled.
+ */
+uint64_t igt_pm_get_wakeup_active_count(struct pci_device *pci_dev)
+{
+ return igt_pm_read_power_attr_u64(pci_dev, "wakeup_active_count");
+}
+
/**
* igt_pm_ignore_slpc_efficient_freq:
* @i915: open i915 drm file descriptor
diff --git a/lib/igt_pm.h b/lib/igt_pm.h
index 12617c2ef..eba54efa2 100644
--- a/lib/igt_pm.h
+++ b/lib/igt_pm.h
@@ -106,6 +106,7 @@ int igt_pm_get_runtime_usage(struct pci_device *pci_dev);
bool igt_pm_has_wakeup_support(struct pci_device *pci_dev);
void igt_pm_set_wakeup_enabled(struct pci_device *pci_dev, bool enable);
void igt_pm_restore_wakeup(void);
+uint64_t igt_pm_get_wakeup_active_count(struct pci_device *pci_dev);
void igt_pm_ignore_slpc_efficient_freq(int i915, int gtfd, bool val);
bool igt_has_pci_pm_capability(struct pci_device *pci_dev);
bool igt_pm_pci_pme_supported(struct pci_device *pci_dev,
--
2.53.0
next prev parent reply other threads:[~2026-09-07 14:15 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-07 14:22 [PATCH i-g-t 0/7] Validate PM_PME signalling on display hotplug Pranay Samala
2026-09-07 14:22 ` [PATCH i-g-t 1/7] lib/igt_pci: Add PCI Power Management capability register layout Pranay Samala
2026-09-07 14:22 ` [PATCH i-g-t 2/7] lib/igt_pm: Add PCI PME capability and D state accessors Pranay Samala
2026-09-07 14:22 ` [PATCH i-g-t 3/7] lib/igt_pm: Factor out power attribute path construction Pranay Samala
2026-09-07 14:22 ` [PATCH i-g-t 4/7] lib/igt_pm: Add power/wakeup accessors Pranay Samala
2026-09-07 14:22 ` Pranay Samala [this message]
2026-09-07 14:22 ` [PATCH i-g-t 6/7] lib/igt_pm: Add drm_kms_helper.poll save/restore helpers Pranay Samala
2026-09-07 14:22 ` [PATCH i-g-t 7/7] tests/chamelium/kms_chamelium_hpd: Add HPD from runtime suspended D3hot Pranay Samala
2026-09-07 19:49 ` ✓ Xe.CI.BAT: success for Validate PM_PME signalling on display hotplug (rev2) Patchwork
2026-09-07 20:02 ` ✓ i915.CI.BAT: " Patchwork
2026-09-08 0:17 ` ✗ Xe.CI.FULL: failure " Patchwork
2026-09-08 6:28 ` ✗ i915.CI.Full: " 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=20260907142259.750528-6-pranay.samala@intel.com \
--to=pranay.samala@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=karthik.b.s@intel.com \
--cc=sameer.lattannavar@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