From cda7ac8ce7de84cf32a3871ba5f318aa3b79381e Mon Sep 17 00:00:00 2001 From: "Gautham R. Shenoy" Date: Thu, 12 Jun 2025 17:53:54 +0530 Subject: [PATCH 1/8] pm: cpupower: Fix the snapshot-order of tsc,mperf, clock in mperf_stop() In the function mperf_start(), mperf_monitor snapshots the time, tsc and finally the aperf,mperf MSRs. However, this order of snapshotting in is reversed in mperf_stop(). As a result, the C0 residency (which is computed as delta_mperf * 100 / delta_tsc) is under-reported on CPUs that is 100% busy. Fix this by snapshotting time, tsc and then aperf,mperf in mperf_stop() in the same order as in mperf_start(). Link: https://lore.kernel.org/r/20250612122355.19629-2-gautham.shenoy@amd.com Signed-off-by: Gautham R. Shenoy Signed-off-by: Shuah Khan --- tools/power/cpupower/utils/idle_monitor/mperf_monitor.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/power/cpupower/utils/idle_monitor/mperf_monitor.c b/tools/power/cpupower/utils/idle_monitor/mperf_monitor.c index 73b6b10cbdd2..5ae02c3d5b64 100644 --- a/tools/power/cpupower/utils/idle_monitor/mperf_monitor.c +++ b/tools/power/cpupower/utils/idle_monitor/mperf_monitor.c @@ -240,9 +240,9 @@ static int mperf_stop(void) int cpu; for (cpu = 0; cpu < cpu_count; cpu++) { - mperf_measure_stats(cpu); - mperf_get_tsc(&tsc_at_measure_end[cpu]); clock_gettime(CLOCK_REALTIME, &time_end[cpu]); + mperf_get_tsc(&tsc_at_measure_end[cpu]); + mperf_measure_stats(cpu); } return 0; From 14a3318b4ac8ae0ca2e1132a89de167e1030fbdb Mon Sep 17 00:00:00 2001 From: "Gautham R. Shenoy" Date: Thu, 12 Jun 2025 17:53:55 +0530 Subject: [PATCH 2/8] pm: cpupower: Fix printing of CORE, CPU fields in cpupower-monitor After the commit 0014f65e3df0 ("pm: cpupower: remove hard-coded topology depth values"), "cpupower monitor" output ceased to print the CORE and the CPU fields on a multi-socket platform. The reason for this is that the patch changed the behaviour to break out of the switch-case after printing the PKG details, while prior to the patch, the CORE and the CPU details would also get printed since the "if" condition check would pass for any level whose topology depth was lesser than that of a package. Fix this ensuring all the details below a desired topology depth are printed in the cpupower monitor output. Link: https://lore.kernel.org/r/20250612122355.19629-3-gautham.shenoy@amd.com Fixes: 0014f65e3df0 ("pm: cpupower: remove hard-coded topology depth values") Signed-off-by: Gautham R. Shenoy Signed-off-by: Shuah Khan --- tools/power/cpupower/utils/idle_monitor/cpupower-monitor.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/tools/power/cpupower/utils/idle_monitor/cpupower-monitor.c b/tools/power/cpupower/utils/idle_monitor/cpupower-monitor.c index ad493157f826..e8b3841d5c0f 100644 --- a/tools/power/cpupower/utils/idle_monitor/cpupower-monitor.c +++ b/tools/power/cpupower/utils/idle_monitor/cpupower-monitor.c @@ -121,10 +121,8 @@ void print_header(int topology_depth) switch (topology_depth) { case TOPOLOGY_DEPTH_PKG: printf(" PKG|"); - break; case TOPOLOGY_DEPTH_CORE: printf("CORE|"); - break; case TOPOLOGY_DEPTH_CPU: printf(" CPU|"); break; @@ -167,10 +165,8 @@ void print_results(int topology_depth, int cpu) switch (topology_depth) { case TOPOLOGY_DEPTH_PKG: printf("%4d|", cpu_top.core_info[cpu].pkg); - break; case TOPOLOGY_DEPTH_CORE: printf("%4d|", cpu_top.core_info[cpu].core); - break; case TOPOLOGY_DEPTH_CPU: printf("%4d|", cpu_top.core_info[cpu].cpu); break; From b74710eaff314d6afe4fb0bbe9bc7657bf226fd4 Mon Sep 17 00:00:00 2001 From: "John B. Wyatt IV" Date: Tue, 24 Jun 2025 16:41:04 -0400 Subject: [PATCH 3/8] cpupower: Improve Python binding's Makefile Add a few build variables to make it easier for distributions to package the bindings. Allow current variables to be overwritten by environment variables that are passed to make. CCing Thorsten Leemhuis . https://lore.kernel.org/r/20250624204105.457971-1-jwyatt@redhat.com Signed-off-by: John B. Wyatt IV Signed-off-by: John B. Wyatt IV Signed-off-by: Shuah Khan --- tools/power/cpupower/bindings/python/Makefile | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/tools/power/cpupower/bindings/python/Makefile b/tools/power/cpupower/bindings/python/Makefile index 81db39a03efb..4527cd732b42 100644 --- a/tools/power/cpupower/bindings/python/Makefile +++ b/tools/power/cpupower/bindings/python/Makefile @@ -4,20 +4,22 @@ # This Makefile expects you have already run `make install-lib` in the lib # directory for the bindings to be created. -CC := gcc +CC ?= gcc +# CFLAGS ?= +LDFLAGS ?= -lcpupower HAVE_SWIG := $(shell if which swig >/dev/null 2>&1; then echo 1; else echo 0; fi) HAVE_PYCONFIG := $(shell if which python-config >/dev/null 2>&1; then echo 1; else echo 0; fi) -PY_INCLUDE = $(firstword $(shell python-config --includes)) -INSTALL_DIR = $(shell python3 -c "import site; print(site.getsitepackages()[0])") +PY_INCLUDE ?= $(firstword $(shell python-config --includes)) +INSTALL_DIR ?= $(shell python3 -c "import site; print(site.getsitepackages()[0])") all: _raw_pylibcpupower.so _raw_pylibcpupower.so: raw_pylibcpupower_wrap.o - $(CC) -shared -lcpupower raw_pylibcpupower_wrap.o -o _raw_pylibcpupower.so + $(CC) -shared $(LDFLAGS) raw_pylibcpupower_wrap.o -o _raw_pylibcpupower.so raw_pylibcpupower_wrap.o: raw_pylibcpupower_wrap.c - $(CC) -fPIC -c raw_pylibcpupower_wrap.c $(PY_INCLUDE) + $(CC) $(CFLAGS) $(PY_INCLUDE) -fPIC -c raw_pylibcpupower_wrap.c raw_pylibcpupower_wrap.c: raw_pylibcpupower.swg ifeq ($(HAVE_SWIG),0) From 9047685cfd2911c36ce89a16270aafa71057c507 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Thu, 26 Jun 2025 18:42:44 +0300 Subject: [PATCH 4/8] PM: Don't use "proxy" headers Update header inclusions to follow IWYU (Include What You Use) principle. Signed-off-by: Andy Shevchenko Link: https://patch.msgid.link/20250626154244.324265-1-andriy.shevchenko@linux.intel.com Signed-off-by: Rafael J. Wysocki --- include/linux/pm.h | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/include/linux/pm.h b/include/linux/pm.h index f0bd8fbae4f2..938b1b446a5d 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -8,14 +8,15 @@ #ifndef _LINUX_PM_H #define _LINUX_PM_H -#include -#include -#include -#include -#include -#include -#include #include +#include +#include +#include +#include +#include +#include +#include +#include /* * Callbacks for platform drivers to implement. From d42c7c6fd66a6e2a78ae1da666c5df6c2fde8389 Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Thu, 3 Jul 2025 14:27:06 +0300 Subject: [PATCH 5/8] PM: domains: Add flags to specify power on attach/detach Calling dev_pm_domain_attach()/dev_pm_domain_detach() in bus driver probe/remove functions can affect system behavior when the drivers attached to the bus use devres-managed resources. Since devres actions may need to access device registers, calling dev_pm_domain_detach() too early, i.e., before these actions complete, can cause failures on some systems. One such example is Renesas RZ/G3S SoC-based platforms. If the device clocks are managed via PM domains, invoking dev_pm_domain_detach() in the bus driver's remove function removes the device's clocks from the PM domain, preventing any subsequent pm_runtime_resume*() calls from enabling those clocks. The second argument of dev_pm_domain_attach() specifies whether the PM domain should be powered on during attachment. Likewise, the second argument of dev_pm_domain_detach() indicates whether the domain should be powered off during detachment. Upcoming changes address the issue described above (initially for the platform bus only) by deferring the call to dev_pm_domain_detach() until after devres_release_all() in device_unbind_cleanup(). The detach_power_off field in struct dev_pm_info stores the detach power off info from the second argument of dev_pm_domain_attach(). Because there are cases where the device's PM domain power-on/off behavior must be conditional (e.g., in i2c_device_probe()), the patch introduces PD_FLAG_ATTACH_POWER_ON and PD_FLAG_DETACH_POWER_OFF flags to be passed to dev_pm_domain_attach(). Finally, dev_pm_domain_attach() and its users are updated to use the newly introduced PD_FLAG_ATTACH_POWER_ON and PD_FLAG_DETACH_POWER_OFF macros. This change is preparatory. Signed-off-by: Claudiu Beznea Reviewed-by: Mathieu Poirier Acked-by: Wolfram Sang # I2C Reviewed-by: Ulf Hansson Link: https://patch.msgid.link/20250703112708.1621607-2-claudiu.beznea.uj@bp.renesas.com [ rjw: Changelog adjustments ] Signed-off-by: Rafael J. Wysocki --- drivers/amba/bus.c | 4 ++-- drivers/base/auxiliary.c | 2 +- drivers/base/platform.c | 2 +- drivers/base/power/common.c | 6 +++--- drivers/clk/qcom/apcs-sdx55.c | 2 +- drivers/gpu/drm/display/drm_dp_aux_bus.c | 2 +- drivers/i2c/i2c-core-base.c | 2 +- drivers/mmc/core/sdio_bus.c | 2 +- drivers/rpmsg/rpmsg_core.c | 2 +- drivers/soundwire/bus_type.c | 2 +- drivers/spi/spi.c | 2 +- drivers/tty/serdev/core.c | 2 +- include/linux/pm_domain.h | 10 ++++++++-- 13 files changed, 23 insertions(+), 17 deletions(-) diff --git a/drivers/amba/bus.c b/drivers/amba/bus.c index 71482d639a6d..74e34a07ef72 100644 --- a/drivers/amba/bus.c +++ b/drivers/amba/bus.c @@ -138,7 +138,7 @@ static int amba_read_periphid(struct amba_device *dev) void __iomem *tmp; int i, ret; - ret = dev_pm_domain_attach(&dev->dev, true); + ret = dev_pm_domain_attach(&dev->dev, PD_FLAG_ATTACH_POWER_ON); if (ret) { dev_dbg(&dev->dev, "can't get PM domain: %d\n", ret); goto err_out; @@ -291,7 +291,7 @@ static int amba_probe(struct device *dev) if (ret < 0) break; - ret = dev_pm_domain_attach(dev, true); + ret = dev_pm_domain_attach(dev, PD_FLAG_ATTACH_POWER_ON); if (ret) break; diff --git a/drivers/base/auxiliary.c b/drivers/base/auxiliary.c index dba7c8e13a53..44cd3f85b659 100644 --- a/drivers/base/auxiliary.c +++ b/drivers/base/auxiliary.c @@ -217,7 +217,7 @@ static int auxiliary_bus_probe(struct device *dev) struct auxiliary_device *auxdev = to_auxiliary_dev(dev); int ret; - ret = dev_pm_domain_attach(dev, true); + ret = dev_pm_domain_attach(dev, PD_FLAG_ATTACH_POWER_ON); if (ret) { dev_warn(dev, "Failed to attach to PM Domain : %d\n", ret); return ret; diff --git a/drivers/base/platform.c b/drivers/base/platform.c index 075ec1d1b73a..df1ec34fdf56 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -1396,7 +1396,7 @@ static int platform_probe(struct device *_dev) if (ret < 0) return ret; - ret = dev_pm_domain_attach(_dev, true); + ret = dev_pm_domain_attach(_dev, PD_FLAG_ATTACH_POWER_ON); if (ret) goto out; diff --git a/drivers/base/power/common.c b/drivers/base/power/common.c index 781968a128ff..fecb85fa85ac 100644 --- a/drivers/base/power/common.c +++ b/drivers/base/power/common.c @@ -83,7 +83,7 @@ EXPORT_SYMBOL_GPL(dev_pm_put_subsys_data); /** * dev_pm_domain_attach - Attach a device to its PM domain. * @dev: Device to attach. - * @power_on: Used to indicate whether we should power on the device. + * @flags: indicate whether we should power on/off the device on attach/detach * * The @dev may only be attached to a single PM domain. By iterating through * the available alternatives we try to find a valid PM domain for the device. @@ -100,14 +100,14 @@ EXPORT_SYMBOL_GPL(dev_pm_put_subsys_data); * Returns 0 on successfully attached PM domain, or when it is found that the * device doesn't need a PM domain, else a negative error code. */ -int dev_pm_domain_attach(struct device *dev, bool power_on) +int dev_pm_domain_attach(struct device *dev, u32 flags) { int ret; if (dev->pm_domain) return 0; - ret = acpi_dev_pm_attach(dev, power_on); + ret = acpi_dev_pm_attach(dev, !!(flags & PD_FLAG_ATTACH_POWER_ON)); if (!ret) ret = genpd_dev_pm_attach(dev); diff --git a/drivers/clk/qcom/apcs-sdx55.c b/drivers/clk/qcom/apcs-sdx55.c index 3ba01622d8f0..90dd1f1855c2 100644 --- a/drivers/clk/qcom/apcs-sdx55.c +++ b/drivers/clk/qcom/apcs-sdx55.c @@ -111,7 +111,7 @@ static int qcom_apcs_sdx55_clk_probe(struct platform_device *pdev) * driver, there seems to be no better place to do this. So do it here! */ cpu_dev = get_cpu_device(0); - ret = dev_pm_domain_attach(cpu_dev, true); + ret = dev_pm_domain_attach(cpu_dev, PD_FLAG_ATTACH_POWER_ON); if (ret) { dev_err_probe(dev, ret, "can't get PM domain: %d\n", ret); goto err; diff --git a/drivers/gpu/drm/display/drm_dp_aux_bus.c b/drivers/gpu/drm/display/drm_dp_aux_bus.c index ec7eac6b595f..718c9122bc3a 100644 --- a/drivers/gpu/drm/display/drm_dp_aux_bus.c +++ b/drivers/gpu/drm/display/drm_dp_aux_bus.c @@ -57,7 +57,7 @@ static int dp_aux_ep_probe(struct device *dev) container_of(aux_ep, struct dp_aux_ep_device_with_data, aux_ep); int ret; - ret = dev_pm_domain_attach(dev, true); + ret = dev_pm_domain_attach(dev, PD_FLAG_ATTACH_POWER_ON); if (ret) return dev_err_probe(dev, ret, "Failed to attach to PM Domain\n"); diff --git a/drivers/i2c/i2c-core-base.c b/drivers/i2c/i2c-core-base.c index 2ad2b1838f0f..38eabf1173da 100644 --- a/drivers/i2c/i2c-core-base.c +++ b/drivers/i2c/i2c-core-base.c @@ -573,7 +573,7 @@ static int i2c_device_probe(struct device *dev) goto err_clear_wakeup_irq; do_power_on = !i2c_acpi_waive_d0_probe(dev); - status = dev_pm_domain_attach(&client->dev, do_power_on); + status = dev_pm_domain_attach(&client->dev, do_power_on ? PD_FLAG_ATTACH_POWER_ON : 0); if (status) goto err_clear_wakeup_irq; diff --git a/drivers/mmc/core/sdio_bus.c b/drivers/mmc/core/sdio_bus.c index b66b637e2d57..656601754966 100644 --- a/drivers/mmc/core/sdio_bus.c +++ b/drivers/mmc/core/sdio_bus.c @@ -161,7 +161,7 @@ static int sdio_bus_probe(struct device *dev) if (!id) return -ENODEV; - ret = dev_pm_domain_attach(dev, false); + ret = dev_pm_domain_attach(dev, 0); if (ret) return ret; diff --git a/drivers/rpmsg/rpmsg_core.c b/drivers/rpmsg/rpmsg_core.c index 6ee36adcbdba..bece5e635ee9 100644 --- a/drivers/rpmsg/rpmsg_core.c +++ b/drivers/rpmsg/rpmsg_core.c @@ -479,7 +479,7 @@ static int rpmsg_dev_probe(struct device *dev) struct rpmsg_endpoint *ept = NULL; int err; - err = dev_pm_domain_attach(dev, true); + err = dev_pm_domain_attach(dev, PD_FLAG_ATTACH_POWER_ON); if (err) goto out; diff --git a/drivers/soundwire/bus_type.c b/drivers/soundwire/bus_type.c index 75d6f16efced..bc1e653080d9 100644 --- a/drivers/soundwire/bus_type.c +++ b/drivers/soundwire/bus_type.c @@ -101,7 +101,7 @@ static int sdw_drv_probe(struct device *dev) /* * attach to power domain but don't turn on (last arg) */ - ret = dev_pm_domain_attach(dev, false); + ret = dev_pm_domain_attach(dev, 0); if (ret) return ret; diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 1bc0fdbb1bd7..8200b47b2295 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -427,7 +427,7 @@ static int spi_probe(struct device *dev) if (spi->irq < 0) spi->irq = 0; - ret = dev_pm_domain_attach(dev, true); + ret = dev_pm_domain_attach(dev, PD_FLAG_ATTACH_POWER_ON); if (ret) return ret; diff --git a/drivers/tty/serdev/core.c b/drivers/tty/serdev/core.c index 0213381fa358..d16c207a1a9b 100644 --- a/drivers/tty/serdev/core.c +++ b/drivers/tty/serdev/core.c @@ -399,7 +399,7 @@ static int serdev_drv_probe(struct device *dev) const struct serdev_device_driver *sdrv = to_serdev_device_driver(dev->driver); int ret; - ret = dev_pm_domain_attach(dev, true); + ret = dev_pm_domain_attach(dev, PD_FLAG_ATTACH_POWER_ON); if (ret) return ret; diff --git a/include/linux/pm_domain.h b/include/linux/pm_domain.h index 0b18160901a2..62a35a78ce9b 100644 --- a/include/linux/pm_domain.h +++ b/include/linux/pm_domain.h @@ -36,10 +36,16 @@ * isn't specified, the index just follows the * index for the attached PM domain. * + * PD_FLAG_ATTACH_POWER_ON: Power on the domain during attach. + * + * PD_FLAG_DETACH_POWER_OFF: Power off the domain during detach. + * */ #define PD_FLAG_NO_DEV_LINK BIT(0) #define PD_FLAG_DEV_LINK_ON BIT(1) #define PD_FLAG_REQUIRED_OPP BIT(2) +#define PD_FLAG_ATTACH_POWER_ON BIT(3) +#define PD_FLAG_DETACH_POWER_OFF BIT(4) struct dev_pm_domain_attach_data { const char * const *pd_names; @@ -501,7 +507,7 @@ struct generic_pm_domain *of_genpd_remove_last(struct device_node *np) #endif /* CONFIG_PM_GENERIC_DOMAINS_OF */ #ifdef CONFIG_PM -int dev_pm_domain_attach(struct device *dev, bool power_on); +int dev_pm_domain_attach(struct device *dev, u32 flags); struct device *dev_pm_domain_attach_by_id(struct device *dev, unsigned int index); struct device *dev_pm_domain_attach_by_name(struct device *dev, @@ -518,7 +524,7 @@ int dev_pm_domain_start(struct device *dev); void dev_pm_domain_set(struct device *dev, struct dev_pm_domain *pd); int dev_pm_domain_set_performance_state(struct device *dev, unsigned int state); #else -static inline int dev_pm_domain_attach(struct device *dev, bool power_on) +static inline int dev_pm_domain_attach(struct device *dev, u32 flags) { return 0; } From f99508074e78fea17f06d753d9ef453b174ec98e Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Thu, 3 Jul 2025 14:27:07 +0300 Subject: [PATCH 6/8] PM: domains: Detach on device_unbind_cleanup() The dev_pm_domain_attach() function is typically used in bus code alongside dev_pm_domain_detach(), often following patterns like: static int bus_probe(struct device *_dev) { struct bus_driver *drv = to_bus_driver(dev->driver); struct bus_device *dev = to_bus_device(_dev); int ret; // ... ret = dev_pm_domain_attach(_dev, true); if (ret) return ret; if (drv->probe) ret = drv->probe(dev); // ... } static void bus_remove(struct device *_dev) { struct bus_driver *drv = to_bus_driver(dev->driver); struct bus_device *dev = to_bus_device(_dev); if (drv->remove) drv->remove(dev); dev_pm_domain_detach(_dev); } When the driver's probe function uses devres-managed resources that depend on the power domain state, those resources are released later during device_unbind_cleanup(). Releasing devres-managed resources that depend on the power domain state after detaching the device from its PM domain can cause failures. For example, if the driver uses devm_pm_runtime_enable() in its probe function, and the device's clocks are managed by the PM domain, then during removal the runtime PM is disabled in device_unbind_cleanup() after the clocks have been removed from the PM domain. It may happen that the devm_pm_runtime_enable() action causes the device to be runtime- resumed. If the driver specific runtime PM APIs access registers directly, this will lead to accessing device registers without clocks being enabled. Similar issues may occur with other devres actions that access device registers. Add detach_power_off member to struct dev_pm_info, to be used later in device_unbind_cleanup() as the power_off argument for dev_pm_domain_detach(). This is a preparatory step toward removing dev_pm_domain_detach() calls from bus remove functions. Since the current PM domain detach functions (genpd_dev_pm_detach() and acpi_dev_pm_detach()) already set dev->pm_domain = NULL, there should be no issues with bus drivers that still call dev_pm_domain_detach() in their remove functions. Signed-off-by: Claudiu Beznea Reviewed-by: Ulf Hansson Link: https://patch.msgid.link/20250703112708.1621607-3-claudiu.beznea.uj@bp.renesas.com Signed-off-by: Rafael J. Wysocki --- drivers/base/dd.c | 2 ++ drivers/base/power/common.c | 3 +++ include/linux/pm.h | 1 + 3 files changed, 6 insertions(+) diff --git a/drivers/base/dd.c b/drivers/base/dd.c index b526e0e0f52d..13ab98e033ea 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include @@ -552,6 +553,7 @@ static void device_unbind_cleanup(struct device *dev) dev->dma_range_map = NULL; device_set_driver(dev, NULL); dev_set_drvdata(dev, NULL); + dev_pm_domain_detach(dev, dev->power.detach_power_off); if (dev->pm_domain && dev->pm_domain->dismiss) dev->pm_domain->dismiss(dev); pm_runtime_reinit(dev); diff --git a/drivers/base/power/common.c b/drivers/base/power/common.c index fecb85fa85ac..6ecf9ce4a4e6 100644 --- a/drivers/base/power/common.c +++ b/drivers/base/power/common.c @@ -111,6 +111,9 @@ int dev_pm_domain_attach(struct device *dev, u32 flags) if (!ret) ret = genpd_dev_pm_attach(dev); + if (dev->pm_domain) + dev->power.detach_power_off = !!(flags & PD_FLAG_DETACH_POWER_OFF); + return ret < 0 ? ret : 0; } EXPORT_SYMBOL_GPL(dev_pm_domain_attach); diff --git a/include/linux/pm.h b/include/linux/pm.h index 938b1b446a5d..14e8370887e3 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -721,6 +721,7 @@ struct dev_pm_info { struct pm_subsys_data *subsys_data; /* Owned by the subsystem. */ void (*set_latency_tolerance)(struct device *, s32); struct dev_pm_qos *qos; + bool detach_power_off:1; /* Owned by the driver core */ }; extern int dev_pm_get_subsys_data(struct device *dev); From ba2ebd52a22eb7306a2093924920a125ad91215a Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Thu, 3 Jul 2025 14:27:08 +0300 Subject: [PATCH 7/8] driver core: platform: Drop dev_pm_domain_detach() call On the Renesas RZ/G3S (and other Renesas SoCs, e.g., RZ/G2{L, LC, UL}), clocks are managed through PM domains. These PM domains, registered on behalf of the clock controller driver, are configured with GENPD_FLAG_PM_CLK. In most of the Renesas drivers used by RZ SoCs, the clocks are enabled/disabled using runtime PM APIs. The power domains may also have power_on/power_off support implemented. After the device PM domain is powered off any CPU accesses to these domains leads to system aborts. During probe, devices are attached to the PM domain controlling their clocks and power. Similarly, during removal, devices are detached from the PM domain. The detachment call stack is as follows: device_driver_detach() -> device_release_driver_internal() -> __device_release_driver() -> device_remove() -> platform_remove() -> dev_pm_domain_detach() During driver unbind, after the device is detached from its PM domain, the device_unbind_cleanup() function is called, which subsequently invokes devres_release_all(). This function handles devres resource cleanup. If runtime PM is enabled in driver probe via devm_pm_runtime_enable(), the cleanup process triggers the action or reset function for disabling runtime PM. This function is pm_runtime_disable_action(), which leads to the following call stack of interest when called: pm_runtime_disable_action() -> pm_runtime_dont_use_autosuspend() -> __pm_runtime_use_autosuspend() -> update_autosuspend() -> rpm_idle() The rpm_idle() function attempts to resume the device at runtime. However, at the point it is called, the device is no longer part of a PM domain (which manages clocks and power states). If the driver implements its own runtime PM APIs for specific functionalities - such as the rzg2l_adc driver - while also relying on the power domain subsystem for power management, rpm_idle() will invoke the driver's runtime PM API. However, since the device is no longer part of a PM domain at this point, the PM domain's runtime PM APIs will not be called. This leads to system aborts on Renesas SoCs. Another identified case is when a subsystem performs various cleanups using device_unbind_cleanup(), calling driver-specific APIs in the process. A known example is the thermal subsystem, which may call driver- specific APIs to disable the thermal device. The relevant call stack in this case is: device_driver_detach() -> device_release_driver_internal() -> device_unbind_cleanup() -> devres_release_all() -> devm_thermal_of_zone_release() -> thermal_zone_device_disable() -> thermal_zone_device_set_mode() -> struct thermal_zone_device_ops::change_mode() At the moment the driver-specific change_mode() API is called, the device is no longer part of its PM domain. Accessing its registers without proper power management leads to system aborts. Drop the call to dev_pm_domain_detach() from the platform bus remove function and rely on the newly introduced call in device_unbind_cleanup(). This ensures the same effect, but the call now occurs after all driver-specific devres resources have been freed. Signed-off-by: Claudiu Beznea Reviewed-by: Ulf Hansson Link: https://patch.msgid.link/20250703112708.1621607-4-claudiu.beznea.uj@bp.renesas.com Signed-off-by: Rafael J. Wysocki --- drivers/base/platform.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/drivers/base/platform.c b/drivers/base/platform.c index df1ec34fdf56..09450349cf32 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -1396,15 +1396,13 @@ static int platform_probe(struct device *_dev) if (ret < 0) return ret; - ret = dev_pm_domain_attach(_dev, PD_FLAG_ATTACH_POWER_ON); + ret = dev_pm_domain_attach(_dev, PD_FLAG_ATTACH_POWER_ON | + PD_FLAG_DETACH_POWER_OFF); if (ret) goto out; - if (drv->probe) { + if (drv->probe) ret = drv->probe(dev); - if (ret) - dev_pm_domain_detach(_dev, true); - } out: if (drv->prevent_deferred_probe && ret == -EPROBE_DEFER) { @@ -1422,7 +1420,6 @@ static void platform_remove(struct device *_dev) if (drv->remove) drv->remove(dev); - dev_pm_domain_detach(_dev, true); } static void platform_shutdown(struct device *_dev) From ebf266d0709b9b8eb3df1fde4152cdd329726598 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 19 Jul 2025 14:40:31 +0200 Subject: [PATCH 8/8] PM: docs: Use my kernel.org address in ABI docs and DT bindings For the sake of consistency, use my kernel.org address in all Contact records in sysfs-devices-power and sysfs-power, and in the power-domain DT binding. Signed-off-by: Rafael J. Wysocki Link: https://patch.msgid.link/5911353.DvuYhMxLoT@rjwysocki.net --- Documentation/ABI/testing/sysfs-devices-power | 34 +++++++++---------- Documentation/ABI/testing/sysfs-power | 26 +++++++------- .../bindings/power/power-domain.yaml | 2 +- 3 files changed, 31 insertions(+), 31 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-devices-power b/Documentation/ABI/testing/sysfs-devices-power index 54195530e97a..9330fb3c79e2 100644 --- a/Documentation/ABI/testing/sysfs-devices-power +++ b/Documentation/ABI/testing/sysfs-devices-power @@ -1,6 +1,6 @@ What: /sys/devices/.../power/ Date: January 2009 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../power directory contains attributes allowing the user space to check and modify some power @@ -8,7 +8,7 @@ Description: What: /sys/devices/.../power/wakeup Date: January 2009 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../power/wakeup attribute allows the user space to check if the device is enabled to wake up the system @@ -34,7 +34,7 @@ Description: What: /sys/devices/.../power/control Date: January 2009 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../power/control attribute allows the user space to control the run-time power management of the device. @@ -53,7 +53,7 @@ Description: What: /sys/devices/.../power/async Date: January 2009 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../async attribute allows the user space to enable or diasble the device's suspend and resume callbacks to @@ -79,7 +79,7 @@ Description: What: /sys/devices/.../power/wakeup_count Date: September 2010 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../wakeup_count attribute contains the number of signaled wakeup events associated with the device. This @@ -90,7 +90,7 @@ Description: What: /sys/devices/.../power/wakeup_active_count Date: September 2010 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../wakeup_active_count attribute contains the number of times the processing of wakeup events associated with @@ -102,7 +102,7 @@ Description: What: /sys/devices/.../power/wakeup_abort_count Date: February 2012 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../wakeup_abort_count attribute contains the number of times the processing of a wakeup event associated with @@ -114,7 +114,7 @@ Description: What: /sys/devices/.../power/wakeup_expire_count Date: February 2012 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../wakeup_expire_count attribute contains the number of times a wakeup event associated with the device has @@ -126,7 +126,7 @@ Description: What: /sys/devices/.../power/wakeup_active Date: September 2010 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../wakeup_active attribute contains either 1, or 0, depending on whether or not a wakeup event associated with @@ -138,7 +138,7 @@ Description: What: /sys/devices/.../power/wakeup_total_time_ms Date: September 2010 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../wakeup_total_time_ms attribute contains the total time of processing wakeup events associated with the @@ -149,7 +149,7 @@ Description: What: /sys/devices/.../power/wakeup_max_time_ms Date: September 2010 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../wakeup_max_time_ms attribute contains the maximum time of processing a single wakeup event associated @@ -161,7 +161,7 @@ Description: What: /sys/devices/.../power/wakeup_last_time_ms Date: September 2010 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../wakeup_last_time_ms attribute contains the value of the monotonic clock corresponding to the time of @@ -173,7 +173,7 @@ Description: What: /sys/devices/.../power/wakeup_prevent_sleep_time_ms Date: February 2012 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../wakeup_prevent_sleep_time_ms attribute contains the total time the device has been preventing @@ -203,7 +203,7 @@ Description: What: /sys/devices/.../power/pm_qos_resume_latency_us Date: March 2012 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../power/pm_qos_resume_latency_us attribute contains the PM QoS resume latency limit for the given device, @@ -223,7 +223,7 @@ Description: What: /sys/devices/.../power/pm_qos_latency_tolerance_us Date: January 2014 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../power/pm_qos_latency_tolerance_us attribute contains the PM QoS active state latency tolerance limit for the @@ -248,7 +248,7 @@ Description: What: /sys/devices/.../power/pm_qos_no_power_off Date: September 2012 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../power/pm_qos_no_power_off attribute is used for manipulating the PM QoS "no power off" flag. If @@ -263,7 +263,7 @@ Description: What: /sys/devices/.../power/runtime_status Date: April 2010 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/devices/.../power/runtime_status attribute contains the current runtime PM status of the device, which may be diff --git a/Documentation/ABI/testing/sysfs-power b/Documentation/ABI/testing/sysfs-power index 2192478e83cf..4d8e1ad020f0 100644 --- a/Documentation/ABI/testing/sysfs-power +++ b/Documentation/ABI/testing/sysfs-power @@ -1,6 +1,6 @@ What: /sys/power/ Date: August 2006 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power directory will contain files that will provide a unified interface to the power management @@ -8,7 +8,7 @@ Description: What: /sys/power/state Date: November 2016 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/state file controls system sleep states. Reading from this file returns the available sleep state @@ -23,7 +23,7 @@ Description: What: /sys/power/mem_sleep Date: November 2016 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/mem_sleep file controls the operating mode of system suspend. Reading from it returns the available modes @@ -41,7 +41,7 @@ Description: What: /sys/power/disk Date: September 2006 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/disk file controls the operating mode of the suspend-to-disk mechanism. Reading from this file returns @@ -90,7 +90,7 @@ Description: What: /sys/power/image_size Date: August 2006 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/image_size file controls the size of the image created by the suspend-to-disk mechanism. It can be written a @@ -107,7 +107,7 @@ Description: What: /sys/power/pm_trace Date: August 2006 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/pm_trace file controls the code which saves the last PM event point in the RTC across reboots, so that you can @@ -156,7 +156,7 @@ Description: What: /sys/power/pm_async Date: January 2009 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/pm_async file controls the switch allowing the user space to enable or disable asynchronous suspend and resume @@ -169,7 +169,7 @@ Description: What: /sys/power/wakeup_count Date: July 2010 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/wakeup_count file allows user space to put the system into a sleep state while taking into account the @@ -184,7 +184,7 @@ Description: What: /sys/power/reserved_size Date: May 2011 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/reserved_size file allows user space to control the amount of memory reserved for allocations made by device @@ -198,7 +198,7 @@ Description: What: /sys/power/autosleep Date: April 2012 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/autosleep file can be written one of the strings returned by reads from /sys/power/state. If that happens, a @@ -215,7 +215,7 @@ Description: What: /sys/power/wake_lock Date: February 2012 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/wake_lock file allows user space to create wakeup source objects and activate them on demand (if one of @@ -242,7 +242,7 @@ Description: What: /sys/power/wake_unlock Date: February 2012 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/wake_unlock file allows user space to deactivate wakeup sources created with the help of /sys/power/wake_lock. @@ -283,7 +283,7 @@ Description: What: /sys/power/pm_debug_messages Date: July 2017 -Contact: Rafael J. Wysocki +Contact: Rafael J. Wysocki Description: The /sys/power/pm_debug_messages file controls the printing of debug messages from the system suspend/hiberbation diff --git a/Documentation/devicetree/bindings/power/power-domain.yaml b/Documentation/devicetree/bindings/power/power-domain.yaml index 8fdb529d560b..b1147dbf2e73 100644 --- a/Documentation/devicetree/bindings/power/power-domain.yaml +++ b/Documentation/devicetree/bindings/power/power-domain.yaml @@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Generic PM domains maintainers: - - Rafael J. Wysocki + - Rafael J. Wysocki - Kevin Hilman - Ulf Hansson