kernel/linux.git/drivers/pmdomain/arm, branch v6.18.21

pmdomain: arm: scmi: Fix genpd leak on provider registration failure

2025-10-22T14:16:14+00:00

If of_genpd_add_provider_onecell() fails during probe, the previously created generic power domains are not removed, leading to a memory leak and potential kernel crash later in genpd_debug_add(). Add proper error handling to unwind the initialized domains before returning from probe to ensure all resources are correctly released on failure. Example crash trace observed without this fix: | Unable to handle kernel paging request at virtual address fffffffffffffc70 | CPU: 1 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.18.0-rc1 #405 PREEMPT | Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno Development Platform | pstate: 00000005 (nzcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) | pc : genpd_debug_add+0x2c/0x160 | lr : genpd_debug_init+0x74/0x98 | Call trace: | genpd_debug_add+0x2c/0x160 (P) | genpd_debug_init+0x74/0x98 | do_one_initcall+0xd0/0x2d8 | do_initcall_level+0xa0/0x140 | do_initcalls+0x60/0xa8 | do_basic_setup+0x28/0x40 | kernel_init_freeable+0xe8/0x170 | kernel_init+0x2c/0x140 | ret_from_fork+0x10/0x20 Fixes: 898216c97ed2 ("firmware: arm_scmi: add device power domain support using genpd") Signed-off-by: Sudeep Holla Reviewed-by: Peng Fan Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson

pmdomain: arm: scmi_pm_domain: remove code clutter

2025-06-18T12:44:43+00:00

There is no need to introduce the boolean power_on to select the constant value for state. Simply pass the value for state as argument. Just remove this code clutter. No functional change. Signed-off-by: Lukas Bulwahn Reviewed-by: Javier Martinez Canillas Reviewed-by: Sudeep Holla Link: https://lore.kernel.org/r/20250606090802.597504-1-lukas.bulwahn@redhat.com Signed-off-by: Ulf Hansson

pmdomain: arm: Do not enable by default during compile testing

2025-04-25T10:15:52+00:00

Enabling the compile test should not cause automatic enabling of all drivers, but only allow to choose to compile them. Signed-off-by: Krzysztof Kozlowski Reviewed-by: Cristian Marussi Link: https://lore.kernel.org/r/20250417074645.81480-1-krzysztof.kozlowski@linaro.org Signed-off-by: Ulf Hansson

pmdomain: arm: scmi_pm_domain: Remove redundant state verification

2025-03-17T10:12:01+00:00

Currently, scmi_pd_power() explicitly verifies whether the requested power state was applied by calling state_get(). While this check could detect failures where the state was not properly updated, ensuring correctness is the responsibility of the SCMI firmware. Removing this redundant state_get() call eliminates an unnecessary round-trip to the firmware, improving efficiency. Any mismatches between the requested and actual states should be handled by the SCMI firmware, which must return a failure if state_set() is unsuccessful. Additionally, in some cases, checking the state after powering off a domain may be unreliable or unsafe, depending on the firmware implementation. This patch removes the redundant verification, simplifying the function without compromising correctness. Reported-and-tested-by: Ranjani Vaidyanathan Signed-off-by: Sudeep Holla Reviewed-by: Peng Fan Reviewed-by: Cristian Marussi Link: https://lore.kernel.org/r/20250314095851.443979-1-sudeep.holla@arm.com Signed-off-by: Ulf Hansson

pmdomain: arm: scmi_pm_domain: Send an explicit request to set the current state

2025-01-17T11:17:16+00:00

On a system with multiple active SCMI agents, one agent(other than OSPM/ Linux or bootloader) would request to turn on a shared power domain before the Linux boots/initialise the genpds. So when the Linux boots and gets the power state as already ON, it just registers the genpd with a default ON state. However, when the driver that needs this shared power domain is probed genpd sees that the power domain status is ON and never makes any SCMI call to power it up which is correct. But, since Linux didn't make an explicit request to turn on the shared power domain, the SCMI platform firmware will not know if the OSPM agent is actively using it. Suppose the other agent that requested the shared power domain to be powered ON requests to power it OFF as it no longer needs it, the SCMI platform firmware needs to turn it off if there are no active users of it which in the above scenaro is the case. As a result of SCMI platform firmware turning off the resource, OSPM/ Linux will crash the moment as it expects the shared power domain to be powered ON. Send an explicit request to set the current state when setting up the genpd power domains so that OSPM registers its vote in the power domain state with the SCMI platform firmware. The other option is to not read the state and set the genpds as default OFF, but it can't handle the scenario on certain platforms where SCMI platform keeps all the power domains turned ON by default for faster boot (or any other such variations) and expect the OSPM to turn off the unused domains if power saving is required. Link: https://lore.kernel.org/all/Z4aBkezSWOPCXcUh@bogus Reported-by: Ranjani Vaidyanathan Reported-by: Peng Fan Signed-off-by: Sudeep Holla Reviewed-by: Cristian Marussi Link: https://lore.kernel.org/r/20250115113931.1181309-1-sudeep.holla@arm.com Signed-off-by: Ulf Hansson

pmdomain: arm: Use FLAG_DEV_NAME_FW to ensure unique names

2024-10-30T16:11:56+00:00

The domain attributes returned by the perf protocol can end up reporting identical names across domains, resulting in debugfs node creation failure. Use the GENPD_FLAG_DEV_NAME_FW to ensure the genpd providers end up with an unique name. Logs: [X1E reports 'NCC' for all its scmi perf domains] debugfs: Directory 'NCC' with parent 'pm_genpd' already present! debugfs: Directory 'NCC' with parent 'pm_genpd' already present! Reported-by: Johan Hovold Closes: https://lore.kernel.org/lkml/ZoQjAWse2YxwyRJv@hovoldconsulting.com/ Suggested-by: Ulf Hansson Suggested-by: Dmitry Baryshkov Signed-off-by: Sibi Sankar Cc: stable@vger.kernel.org Message-ID: <20241030125512.2884761-6-quic_sibis@quicinc.com> Signed-off-by: Ulf Hansson

pmdomain: arm: scmi_pm_domain: set flag GENPD_FLAG_ACTIVE_WAKEUP

2024-06-05T09:54:43+00:00

Set flag GENPD_FLAG_ACTIVE_WAKEUP to scmi genpd, then when a device is set as wakeup source using device_set_wakeup_enable, the power domain could be kept on to make sure the device could wakeup the system. Signed-off-by: Peng Fan Acked-by: Sudeep Holla Link: https://lore.kernel.org/r/20240514131833.911703-1-peng.fan@oss.nxp.com Signed-off-by: Ulf Hansson

pmdomain: arm: Fix NULL dereference on scmi_perf_domain removal

2024-02-13T12:04:19+00:00

On unloading of the scmi_perf_domain module got the below splat, when in the DT provided to the system under test the '#power-domain-cells' property was missing. Indeed, this particular setup causes the probe to bail out early without giving any error, which leads to the ->remove() callback gets to run too, but without all the expected initialized structures in place. Add a check and bail out early on remove too. Call trace: scmi_perf_domain_remove+0x28/0x70 [scmi_perf_domain] scmi_dev_remove+0x28/0x40 [scmi_core] device_remove+0x54/0x90 device_release_driver_internal+0x1dc/0x240 driver_detach+0x58/0xa8 bus_remove_driver+0x78/0x108 driver_unregister+0x38/0x70 scmi_driver_unregister+0x28/0x180 [scmi_core] scmi_perf_domain_driver_exit+0x18/0xb78 [scmi_perf_domain] __arm64_sys_delete_module+0x1a8/0x2c0 invoke_syscall+0x50/0x128 el0_svc_common.constprop.0+0x48/0xf0 do_el0_svc+0x24/0x38 el0_svc+0x34/0xb8 el0t_64_sync_handler+0x100/0x130 el0t_64_sync+0x190/0x198 Code: a90153f3 f9403c14 f9414800 955f8a05 (b9400a80) ---[ end trace 0000000000000000 ]--- Fixes: 2af23ceb8624 ("pmdomain: arm: Add the SCMI performance domain") Signed-off-by: Cristian Marussi Reviewed-by: Sudeep Holla Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20240125191756.868860-1-cristian.marussi@arm.com Signed-off-by: Ulf Hansson

pmdomain: Merge branch fixes into next

2023-11-27T15:28:51+00:00

Merge the pmdomain fixes for v6.7-rc[n] into the next branch, to allow them to get tested together with the changes that are targeted for v6.8. Signed-off-by: Ulf Hansson

pmdomain: arm: Avoid polling for scmi_perf_domain

2023-11-27T15:28:20+00:00

It was a mistake to prefer polling based mode when setting a performance level for a domain. Let's instead rely on the protocol to decide what is best and thus avoid polling when possible. Reported-by: Nikunj Kela Fixes: 2af23ceb8624 ("pmdomain: arm: Add the SCMI performance domain") Signed-off-by: Ulf Hansson Reviewed-by: Sudeep Holla Link: https://lore.kernel.org/r/20231127135033.136442-1-ulf.hansson@linaro.org