linux-stable.git/drivers/clk, branch v6.9.2 Linux kernel stable tree clk: samsung: Revert "clk: Use device_get_match_data()" 2024-04-30T21:24:27+00:00 Marek Szyprowski m.szyprowski@samsung.com 2024-04-30T18:46:56+00:00 aacb99de1099346244d488bdf7df489a44278574 device_get_match_data() function should not be used on the device other than the one matched to the given driver, because it always returns the match_data of the matched driver. In case of exynos-clkout driver, the original code matches the OF IDs on the PARENT device, so replacing it with of_device_get_match_data() broke the driver. This has been already pointed once in commit 2bc5febd05ab ("clk: samsung: Revert "clk: samsung: exynos-clkout: Use of_device_get_match_data()""). To avoid further confusion, add a comment about this special case, which requires direct of_match_device() call to pass custom IDs array. This partially reverts commit 409c39ec92a35e3708f5b5798c78eae78512cd71. Cc: <stable@vger.kernel.org> Fixes: 409c39ec92a3 ("clk: Use device_get_match_data()") Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Link: https://lore.kernel.org/r/20240425075628.838497-1-m.szyprowski@samsung.com Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Link: https://lore.kernel.org/r/20240430184656.357805-1-krzysztof.kozlowski@linaro.org Signed-off-by: Stephen Boyd <sboyd@kernel.org> 
device_get_match_data() function should not be used on the device other
than the one matched to the given driver, because it always returns the
match_data of the matched driver. In case of exynos-clkout driver, the
original code matches the OF IDs on the PARENT device, so replacing it
with of_device_get_match_data() broke the driver.

This has been already pointed once in commit 2bc5febd05ab ("clk: samsung:
Revert "clk: samsung: exynos-clkout: Use of_device_get_match_data()"").
To avoid further confusion, add a comment about this special case, which
requires direct of_match_device() call to pass custom IDs array.

This partially reverts commit 409c39ec92a35e3708f5b5798c78eae78512cd71.

Cc: <stable@vger.kernel.org>
Fixes: 409c39ec92a3 ("clk: Use device_get_match_data()")
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Link: https://lore.kernel.org/r/20240425075628.838497-1-m.szyprowski@samsung.com
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20240430184656.357805-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Merge tag 'sunxi-clk-fixes-for-6.9-1' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux into clk-fixes 2024-04-29T21:59:35+00:00 Stephen Boyd sboyd@kernel.org 2024-04-29T21:59:35+00:00 a22549304372086420d0d5dc99661090e5c388b9 Pull Allwinner clk driver fixes from Jernej Skrabec: - fix H6 CPU rate change via reparenting - set A64 MIPI PLL min & max rate * tag 'sunxi-clk-fixes-for-6.9-1' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux: clk: sunxi-ng: a64: Set minimum and maximum rate for PLL-MIPI clk: sunxi-ng: common: Support minimum and maximum rate clk: sunxi-ng: h6: Reparent CPUX during PLL CPUX rate change 
Pull Allwinner clk driver fixes from Jernej Skrabec:

 - fix H6 CPU rate change via reparenting
 - set A64 MIPI PLL min & max rate

* tag 'sunxi-clk-fixes-for-6.9-1' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
  clk: sunxi-ng: a64: Set minimum and maximum rate for PLL-MIPI
  clk: sunxi-ng: common: Support minimum and maximum rate
  clk: sunxi-ng: h6: Reparent CPUX during PLL CPUX rate change
Merge tag 'qcom-clk-fixes-for-6.9' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux into clk-fixes 2024-04-22T18:59:01+00:00 Stephen Boyd sboyd@kernel.org 2024-04-22T18:59:01+00:00 f8981b0d529513a78d30ae659b74269ae40fa2ab Pull Qualcomm clk driver fixes from Bjorn Andersson: The introduction of support for the external VDD_GFX supply in SA8295P ADP unearthed a lockdep problem in the GDSC code w.r.t regulator supplies. Make the regulator optional, to avoid creating a dummy regulator, on those boards that doesn't use this. While not solving the root cause of the problem, it reduces the impact of the lockdep warning - and it avoids wasting resources. Refactoring of the RPM clock driver accidentally removed num_clks from msm8976. Reintroduce this to get the clocks back. * tag 'qcom-clk-fixes-for-6.9' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux: clk: qcom: smd-rpm: Restore msm8976 num_clk clk: qcom: gdsc: treat optional supplies as optional 
Pull Qualcomm clk driver fixes from Bjorn Andersson:

The introduction of support for the external VDD_GFX supply in SA8295P
ADP unearthed a lockdep problem in the GDSC code w.r.t regulator
supplies. Make the regulator optional, to avoid creating a dummy
regulator, on those boards that doesn't use this. While not solving the
root cause of the problem, it reduces the impact of the lockdep warning
- and it avoids wasting resources.

Refactoring of the RPM clock driver accidentally removed num_clks from
msm8976. Reintroduce this to get the clocks back.

* tag 'qcom-clk-fixes-for-6.9' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux:
  clk: qcom: smd-rpm: Restore msm8976 num_clk
  clk: qcom: gdsc: treat optional supplies as optional
clk: sunxi-ng: a64: Set minimum and maximum rate for PLL-MIPI 2024-04-15T21:21:56+00:00 Frank Oltmanns frank@oltmanns.dev 2024-03-10T13:21:12+00:00 69f16d9b789821183d342719d2ebd4a5ac7178bc When the Allwinner A64's TCON0 searches the ideal rate for the connected panel, it may happen that it requests a rate from its parent PLL-MIPI which PLL-MIPI does not support. This happens for example on the Olimex TERES-I laptop where TCON0 requests PLL-MIPI to change to a rate of several GHz which causes the panel to stay blank. It also happens on the pinephone where a rate of less than 500 MHz is requested which causes instabilities on some phones. Set the minimum and maximum rate of Allwinner A64's PLL-MIPI according to the Allwinner User Manual. Fixes: ca1170b69968 ("clk: sunxi-ng: a64: force select PLL_MIPI in TCON0 mux") Reported-by: Diego Roversi <diegor@tiscali.it> Closes: https://groups.google.com/g/linux-sunxi/c/Rh-Uqqa66bw Tested-by: Diego Roversi <diegor@tiscali.it> Cc: stable@vger.kernel.org Reviewed-by: Maxime Ripard <mripard@kernel.org> Signed-off-by: Frank Oltmanns <frank@oltmanns.dev> Reviewed-by: Jernej Skrabec <jernej.skrabec@gmail.com> Link: https://lore.kernel.org/r/20240310-pinephone-pll-fixes-v4-2-46fc80c83637@oltmanns.dev Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com> 
When the Allwinner A64's TCON0 searches the ideal rate for the connected
panel, it may happen that it requests a rate from its parent PLL-MIPI
which PLL-MIPI does not support.

This happens for example on the Olimex TERES-I laptop where TCON0
requests PLL-MIPI to change to a rate of several GHz which causes the
panel to stay blank. It also happens on the pinephone where a rate of
less than 500 MHz is requested which causes instabilities on some
phones.

Set the minimum and maximum rate of Allwinner A64's PLL-MIPI according
to the Allwinner User Manual.

Fixes: ca1170b69968 ("clk: sunxi-ng: a64: force select PLL_MIPI in TCON0 mux")
Reported-by: Diego Roversi <diegor@tiscali.it>
Closes: https://groups.google.com/g/linux-sunxi/c/Rh-Uqqa66bw
Tested-by: Diego Roversi <diegor@tiscali.it>
Cc: stable@vger.kernel.org
Reviewed-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Frank Oltmanns <frank@oltmanns.dev>
Reviewed-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Link: https://lore.kernel.org/r/20240310-pinephone-pll-fixes-v4-2-46fc80c83637@oltmanns.dev
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
clk: sunxi-ng: common: Support minimum and maximum rate 2024-04-15T21:21:45+00:00 Frank Oltmanns frank@oltmanns.dev 2024-03-10T13:21:11+00:00 b914ec33b391ec766545a41f0cfc0de3e0b388d7 The Allwinner SoC's typically have an upper and lower limit for their clocks' rates. Up until now, support for that has been implemented separately for each clock type. Implement that functionality in the sunxi-ng's common part making use of the CCF rate liming capabilities, so that it is available for all clock types. Suggested-by: Maxime Ripard <mripard@kernel.org> Signed-off-by: Frank Oltmanns <frank@oltmanns.dev> Cc: stable@vger.kernel.org Reviewed-by: Jernej Skrabec <jernej.skrabec@gmail.com> Acked-by: Maxime Ripard <mripard@kernel.org> Link: https://lore.kernel.org/r/20240310-pinephone-pll-fixes-v4-1-46fc80c83637@oltmanns.dev Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com> 
The Allwinner SoC's typically have an upper and lower limit for their
clocks' rates. Up until now, support for that has been implemented
separately for each clock type.

Implement that functionality in the sunxi-ng's common part making use of
the CCF rate liming capabilities, so that it is available for all clock
types.

Suggested-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Frank Oltmanns <frank@oltmanns.dev>
Cc: stable@vger.kernel.org
Reviewed-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Acked-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20240310-pinephone-pll-fixes-v4-1-46fc80c83637@oltmanns.dev
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
clk: sunxi-ng: h6: Reparent CPUX during PLL CPUX rate change 2024-04-15T21:04:22+00:00 Jernej Skrabec jernej.skrabec@gmail.com 2023-10-13T18:17:12+00:00 7e91ed763dc07437777bd012af7a2bd4493731ff While PLL CPUX clock rate change when CPU is running from it works in vast majority of cases, now and then it causes instability. This leads to system crashes and other undefined behaviour. After a lot of testing (30+ hours) while also doing a lot of frequency switches, we can't observe any instability issues anymore when doing reparenting to stable clock like 24 MHz oscillator. Fixes: 524353ea480b ("clk: sunxi-ng: add support for the Allwinner H6 CCU") Reported-by: Chad Wagner <wagnerch42@gmail.com> Link: https://forum.libreelec.tv/thread/27295-orange-pi-3-lts-freezes/ Tested-by: Chad Wagner <wagnerch42@gmail.com> Reviewed-by: Chen-Yu Tsai <wens@csie.org> Link: https://lore.kernel.org/r/20231013181712.2128037-1-jernej.skrabec@gmail.com Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com> 
While PLL CPUX clock rate change when CPU is running from it works in
vast majority of cases, now and then it causes instability. This leads
to system crashes and other undefined behaviour. After a lot of testing
(30+ hours) while also doing a lot of frequency switches, we can't
observe any instability issues anymore when doing reparenting to stable
clock like 24 MHz oscillator.

Fixes: 524353ea480b ("clk: sunxi-ng: add support for the Allwinner H6 CCU")
Reported-by: Chad Wagner <wagnerch42@gmail.com>
Link: https://forum.libreelec.tv/thread/27295-orange-pi-3-lts-freezes/
Tested-by: Chad Wagner <wagnerch42@gmail.com>
Reviewed-by: Chen-Yu Tsai <wens@csie.org>
Link: https://lore.kernel.org/r/20231013181712.2128037-1-jernej.skrabec@gmail.com
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
clk: mediatek: mt7988-infracfg: fix clocks for 2nd PCIe port 2024-04-11T03:50:26+00:00 Daniel Golle daniel@makrotopia.org 2024-03-13T22:05:37+00:00 d3e8a91a848a5941e3c31ecebd6b2612b37e01a6 Due to what seems to be an undocumented oddity in MediaTek's MT7988 SoC design the CLK_INFRA_PCIE_PERI_26M_CK_P2 clock requires CLK_INFRA_PCIE_PERI_26M_CK_P3 to be enabled. This currently leads to PCIe port 2 not working in Linux. Reflect the apparent relationship in the clk driver to make sure PCIe port 2 of the MT7988 SoC works. Fixes: 4b4719437d85f ("clk: mediatek: add drivers for MT7988 SoC") Suggested-by: Sam Shih <sam.shih@mediatek.com> Signed-off-by: Daniel Golle <daniel@makrotopia.org> Link: https://lore.kernel.org/r/1da2506a51f970706bf4ec9509dd04e0471065e5.1710367453.git.daniel@makrotopia.org Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org> 
Due to what seems to be an undocumented oddity in MediaTek's MT7988
SoC design the CLK_INFRA_PCIE_PERI_26M_CK_P2 clock requires
CLK_INFRA_PCIE_PERI_26M_CK_P3 to be enabled.

This currently leads to PCIe port 2 not working in Linux.

Reflect the apparent relationship in the clk driver to make sure PCIe
port 2 of the MT7988 SoC works.

Fixes: 4b4719437d85f ("clk: mediatek: add drivers for MT7988 SoC")
Suggested-by: Sam Shih <sam.shih@mediatek.com>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Link: https://lore.kernel.org/r/1da2506a51f970706bf4ec9509dd04e0471065e5.1710367453.git.daniel@makrotopia.org
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
clk: mediatek: Do a runtime PM get on controllers during probe 2024-04-11T03:50:06+00:00 Pin-yen Lin treapking@chromium.org 2024-03-12T11:51:55+00:00 2f7b1d8b5505efb0057cd1ab85fca206063ea4c3 mt8183-mfgcfg has a mutual dependency with genpd during the probing stage, which leads to a deadlock in the following call stack: CPU0: genpd_lock --> clk_prepare_lock genpd_power_off_work_fn() genpd_lock() generic_pm_domain::power_off() clk_unprepare() clk_prepare_lock() CPU1: clk_prepare_lock --> genpd_lock clk_register() __clk_core_init() clk_prepare_lock() clk_pm_runtime_get() genpd_lock() Do a runtime PM get at the probe function to make sure clk_register() won't acquire the genpd lock. Instead of only modifying mt8183-mfgcfg, do this on all mediatek clock controller probings because we don't believe this would cause any regression. Verified on MT8183 and MT8192 Chromebooks. Fixes: acddfc2c261b ("clk: mediatek: Add MT8183 clock support") Signed-off-by: Pin-yen Lin <treapking@chromium.org> Link: https://lore.kernel.org/r/20240312115249.3341654-1-treapking@chromium.org Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org> 
mt8183-mfgcfg has a mutual dependency with genpd during the probing
stage, which leads to a deadlock in the following call stack:

CPU0:  genpd_lock --> clk_prepare_lock
genpd_power_off_work_fn()
 genpd_lock()
 generic_pm_domain::power_off()
    clk_unprepare()
      clk_prepare_lock()

CPU1: clk_prepare_lock --> genpd_lock
clk_register()
  __clk_core_init()
    clk_prepare_lock()
    clk_pm_runtime_get()
      genpd_lock()

Do a runtime PM get at the probe function to make sure clk_register()
won't acquire the genpd lock. Instead of only modifying mt8183-mfgcfg,
do this on all mediatek clock controller probings because we don't
believe this would cause any regression.

Verified on MT8183 and MT8192 Chromebooks.

Fixes: acddfc2c261b ("clk: mediatek: Add MT8183 clock support")
Signed-off-by: Pin-yen Lin <treapking@chromium.org>

Link: https://lore.kernel.org/r/20240312115249.3341654-1-treapking@chromium.org
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
clk: Get runtime PM before walking tree for clk_summary 2024-04-08T02:31:31+00:00 Stephen Boyd sboyd@kernel.org 2024-03-25T18:41:59+00:00 9d1e795f754db1ac3344528b7af0b17b8146f321 Similar to the previous commit, we should make sure that all devices are runtime resumed before printing the clk_summary through debugfs. Failure to do so would result in a deadlock if the thread is resuming a device to print clk state and that device is also runtime resuming in another thread, e.g the screen is turning on and the display driver is starting up. We remove the calls to clk_pm_runtime_{get,put}() in this path because they're superfluous now that we know the devices are runtime resumed. This also squashes a bug where the return value of clk_pm_runtime_get() wasn't checked, leading to an RPM count underflow on error paths. Fixes: 1bb294a7981c ("clk: Enable/Disable runtime PM for clk_summary") Cc: Taniya Das <quic_tdas@quicinc.com> Cc: Douglas Anderson <dianders@chromium.org> Signed-off-by: Stephen Boyd <sboyd@kernel.org> Link: https://lore.kernel.org/r/20240325184204.745706-6-sboyd@kernel.org Reviewed-by: Douglas Anderson <dianders@chromium.org> 
Similar to the previous commit, we should make sure that all devices are
runtime resumed before printing the clk_summary through debugfs. Failure
to do so would result in a deadlock if the thread is resuming a device
to print clk state and that device is also runtime resuming in another
thread, e.g the screen is turning on and the display driver is starting
up. We remove the calls to clk_pm_runtime_{get,put}() in this path
because they're superfluous now that we know the devices are runtime
resumed. This also squashes a bug where the return value of
clk_pm_runtime_get() wasn't checked, leading to an RPM count underflow
on error paths.

Fixes: 1bb294a7981c ("clk: Enable/Disable runtime PM for clk_summary")
Cc: Taniya Das <quic_tdas@quicinc.com>
Cc: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Link: https://lore.kernel.org/r/20240325184204.745706-6-sboyd@kernel.org
Reviewed-by: Douglas Anderson <dianders@chromium.org>
clk: Get runtime PM before walking tree during disable_unused 2024-04-08T02:31:30+00:00 Stephen Boyd sboyd@kernel.org 2024-03-25T18:41:58+00:00 e581cf5d216289ef292d1a4036d53ce90e122469 Doug reported [1] the following hung task: INFO: task swapper/0:1 blocked for more than 122 seconds. Not tainted 5.15.149-21875-gf795ebc40eb8 #1 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. task:swapper/0 state:D stack: 0 pid: 1 ppid: 0 flags:0x00000008 Call trace: __switch_to+0xf4/0x1f4 __schedule+0x418/0xb80 schedule+0x5c/0x10c rpm_resume+0xe0/0x52c rpm_resume+0x178/0x52c __pm_runtime_resume+0x58/0x98 clk_pm_runtime_get+0x30/0xb0 clk_disable_unused_subtree+0x58/0x208 clk_disable_unused_subtree+0x38/0x208 clk_disable_unused_subtree+0x38/0x208 clk_disable_unused_subtree+0x38/0x208 clk_disable_unused_subtree+0x38/0x208 clk_disable_unused+0x4c/0xe4 do_one_initcall+0xcc/0x2d8 do_initcall_level+0xa4/0x148 do_initcalls+0x5c/0x9c do_basic_setup+0x24/0x30 kernel_init_freeable+0xec/0x164 kernel_init+0x28/0x120 ret_from_fork+0x10/0x20 INFO: task kworker/u16:0:9 blocked for more than 122 seconds. Not tainted 5.15.149-21875-gf795ebc40eb8 #1 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. task:kworker/u16:0 state:D stack: 0 pid: 9 ppid: 2 flags:0x00000008 Workqueue: events_unbound deferred_probe_work_func Call trace: __switch_to+0xf4/0x1f4 __schedule+0x418/0xb80 schedule+0x5c/0x10c schedule_preempt_disabled+0x2c/0x48 __mutex_lock+0x238/0x488 __mutex_lock_slowpath+0x1c/0x28 mutex_lock+0x50/0x74 clk_prepare_lock+0x7c/0x9c clk_core_prepare_lock+0x20/0x44 clk_prepare+0x24/0x30 clk_bulk_prepare+0x40/0xb0 mdss_runtime_resume+0x54/0x1c8 pm_generic_runtime_resume+0x30/0x44 __genpd_runtime_resume+0x68/0x7c genpd_runtime_resume+0x108/0x1f4 __rpm_callback+0x84/0x144 rpm_callback+0x30/0x88 rpm_resume+0x1f4/0x52c rpm_resume+0x178/0x52c __pm_runtime_resume+0x58/0x98 __device_attach+0xe0/0x170 device_initial_probe+0x1c/0x28 bus_probe_device+0x3c/0x9c device_add+0x644/0x814 mipi_dsi_device_register_full+0xe4/0x170 devm_mipi_dsi_device_register_full+0x28/0x70 ti_sn_bridge_probe+0x1dc/0x2c0 auxiliary_bus_probe+0x4c/0x94 really_probe+0xcc/0x2c8 __driver_probe_device+0xa8/0x130 driver_probe_device+0x48/0x110 __device_attach_driver+0xa4/0xcc bus_for_each_drv+0x8c/0xd8 __device_attach+0xf8/0x170 device_initial_probe+0x1c/0x28 bus_probe_device+0x3c/0x9c deferred_probe_work_func+0x9c/0xd8 process_one_work+0x148/0x518 worker_thread+0x138/0x350 kthread+0x138/0x1e0 ret_from_fork+0x10/0x20 The first thread is walking the clk tree and calling clk_pm_runtime_get() to power on devices required to read the clk hardware via struct clk_ops::is_enabled(). This thread holds the clk prepare_lock, and is trying to runtime PM resume a device, when it finds that the device is in the process of resuming so the thread schedule()s away waiting for the device to finish resuming before continuing. The second thread is runtime PM resuming the same device, but the runtime resume callback is calling clk_prepare(), trying to grab the prepare_lock waiting on the first thread. This is a classic ABBA deadlock. To properly fix the deadlock, we must never runtime PM resume or suspend a device with the clk prepare_lock held. Actually doing that is near impossible today because the global prepare_lock would have to be dropped in the middle of the tree, the device runtime PM resumed/suspended, and then the prepare_lock grabbed again to ensure consistency of the clk tree topology. If anything changes with the clk tree in the meantime, we've lost and will need to start the operation all over again. Luckily, most of the time we're simply incrementing or decrementing the runtime PM count on an active device, so we don't have the chance to schedule away with the prepare_lock held. Let's fix this immediate problem that can be triggered more easily by simply booting on Qualcomm sc7180. Introduce a list of clk_core structures that have been registered, or are in the process of being registered, that require runtime PM to operate. Iterate this list and call clk_pm_runtime_get() on each of them without holding the prepare_lock during clk_disable_unused(). This way we can be certain that the runtime PM state of the devices will be active and resumed so we can't schedule away while walking the clk tree with the prepare_lock held. Similarly, call clk_pm_runtime_put() without the prepare_lock held to properly drop the runtime PM reference. We remove the calls to clk_pm_runtime_{get,put}() in this path because they're superfluous now that we know the devices are runtime resumed. Reported-by: Douglas Anderson <dianders@chromium.org> Closes: https://lore.kernel.org/all/20220922084322.RFC.2.I375b6b9e0a0a5348962f004beb3dafee6a12dfbb@changeid/ [1] Closes: https://issuetracker.google.com/328070191 Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Ulf Hansson <ulf.hansson@linaro.org> Cc: Krzysztof Kozlowski <krzk@kernel.org> Fixes: 9a34b45397e5 ("clk: Add support for runtime PM") Signed-off-by: Stephen Boyd <sboyd@kernel.org> Link: https://lore.kernel.org/r/20240325184204.745706-5-sboyd@kernel.org Reviewed-by: Douglas Anderson <dianders@chromium.org> 
Doug reported [1] the following hung task:

 INFO: task swapper/0:1 blocked for more than 122 seconds.
       Not tainted 5.15.149-21875-gf795ebc40eb8 #1
 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
 task:swapper/0       state:D stack:    0 pid:    1 ppid:     0 flags:0x00000008
 Call trace:
  __switch_to+0xf4/0x1f4
  __schedule+0x418/0xb80
  schedule+0x5c/0x10c
  rpm_resume+0xe0/0x52c
  rpm_resume+0x178/0x52c
  __pm_runtime_resume+0x58/0x98
  clk_pm_runtime_get+0x30/0xb0
  clk_disable_unused_subtree+0x58/0x208
  clk_disable_unused_subtree+0x38/0x208
  clk_disable_unused_subtree+0x38/0x208
  clk_disable_unused_subtree+0x38/0x208
  clk_disable_unused_subtree+0x38/0x208
  clk_disable_unused+0x4c/0xe4
  do_one_initcall+0xcc/0x2d8
  do_initcall_level+0xa4/0x148
  do_initcalls+0x5c/0x9c
  do_basic_setup+0x24/0x30
  kernel_init_freeable+0xec/0x164
  kernel_init+0x28/0x120
  ret_from_fork+0x10/0x20
 INFO: task kworker/u16:0:9 blocked for more than 122 seconds.
       Not tainted 5.15.149-21875-gf795ebc40eb8 #1
 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
 task:kworker/u16:0   state:D stack:    0 pid:    9 ppid:     2 flags:0x00000008
 Workqueue: events_unbound deferred_probe_work_func
 Call trace:
  __switch_to+0xf4/0x1f4
  __schedule+0x418/0xb80
  schedule+0x5c/0x10c
  schedule_preempt_disabled+0x2c/0x48
  __mutex_lock+0x238/0x488
  __mutex_lock_slowpath+0x1c/0x28
  mutex_lock+0x50/0x74
  clk_prepare_lock+0x7c/0x9c
  clk_core_prepare_lock+0x20/0x44
  clk_prepare+0x24/0x30
  clk_bulk_prepare+0x40/0xb0
  mdss_runtime_resume+0x54/0x1c8
  pm_generic_runtime_resume+0x30/0x44
  __genpd_runtime_resume+0x68/0x7c
  genpd_runtime_resume+0x108/0x1f4
  __rpm_callback+0x84/0x144
  rpm_callback+0x30/0x88
  rpm_resume+0x1f4/0x52c
  rpm_resume+0x178/0x52c
  __pm_runtime_resume+0x58/0x98
  __device_attach+0xe0/0x170
  device_initial_probe+0x1c/0x28
  bus_probe_device+0x3c/0x9c
  device_add+0x644/0x814
  mipi_dsi_device_register_full+0xe4/0x170
  devm_mipi_dsi_device_register_full+0x28/0x70
  ti_sn_bridge_probe+0x1dc/0x2c0
  auxiliary_bus_probe+0x4c/0x94
  really_probe+0xcc/0x2c8
  __driver_probe_device+0xa8/0x130
  driver_probe_device+0x48/0x110
  __device_attach_driver+0xa4/0xcc
  bus_for_each_drv+0x8c/0xd8
  __device_attach+0xf8/0x170
  device_initial_probe+0x1c/0x28
  bus_probe_device+0x3c/0x9c
  deferred_probe_work_func+0x9c/0xd8
  process_one_work+0x148/0x518
  worker_thread+0x138/0x350
  kthread+0x138/0x1e0
  ret_from_fork+0x10/0x20

The first thread is walking the clk tree and calling
clk_pm_runtime_get() to power on devices required to read the clk
hardware via struct clk_ops::is_enabled(). This thread holds the clk
prepare_lock, and is trying to runtime PM resume a device, when it finds
that the device is in the process of resuming so the thread schedule()s
away waiting for the device to finish resuming before continuing. The
second thread is runtime PM resuming the same device, but the runtime
resume callback is calling clk_prepare(), trying to grab the
prepare_lock waiting on the first thread.

This is a classic ABBA deadlock. To properly fix the deadlock, we must
never runtime PM resume or suspend a device with the clk prepare_lock
held. Actually doing that is near impossible today because the global
prepare_lock would have to be dropped in the middle of the tree, the
device runtime PM resumed/suspended, and then the prepare_lock grabbed
again to ensure consistency of the clk tree topology. If anything
changes with the clk tree in the meantime, we've lost and will need to
start the operation all over again.

Luckily, most of the time we're simply incrementing or decrementing the
runtime PM count on an active device, so we don't have the chance to
schedule away with the prepare_lock held. Let's fix this immediate
problem that can be triggered more easily by simply booting on Qualcomm
sc7180.

Introduce a list of clk_core structures that have been registered, or
are in the process of being registered, that require runtime PM to
operate. Iterate this list and call clk_pm_runtime_get() on each of them
without holding the prepare_lock during clk_disable_unused(). This way
we can be certain that the runtime PM state of the devices will be
active and resumed so we can't schedule away while walking the clk tree
with the prepare_lock held. Similarly, call clk_pm_runtime_put() without
the prepare_lock held to properly drop the runtime PM reference. We
remove the calls to clk_pm_runtime_{get,put}() in this path because
they're superfluous now that we know the devices are runtime resumed.

Reported-by: Douglas Anderson <dianders@chromium.org>
Closes: https://lore.kernel.org/all/20220922084322.RFC.2.I375b6b9e0a0a5348962f004beb3dafee6a12dfbb@changeid/ [1]
Closes: https://issuetracker.google.com/328070191
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Ulf Hansson <ulf.hansson@linaro.org>
Cc: Krzysztof Kozlowski <krzk@kernel.org>
Fixes: 9a34b45397e5 ("clk: Add support for runtime PM")
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Link: https://lore.kernel.org/r/20240325184204.745706-5-sboyd@kernel.org
Reviewed-by: Douglas Anderson <dianders@chromium.org>