Skip to contentSkip to conversation

soc: qcom: update config dependencies for QCOM_RPMPD

PATCH In review Rajendra Nayak 5 participants 11 messages Jan 18, 2019 – Jan 31, 2019

Subsystems (6)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 soc: qcom: update config dependencies for QCOM_RPMPD

Since QCOM_RPMPD is bool and it depends on QCOM_SMD_RPM which is tristate, configurations such as arm64:allmodconfig result in

CONFIG_QCOM_RPMPD=y
CONFIG_QCOM_SMD_RPM=m

This in turn results in

drivers/soc/qcom/rpmpd.o: In function `rpmpd_send_corner':
rpmpd.c:(.text+0x10c): undefined reference to qcom_rpm_smd_write' drivers/soc/qcom/rpmpd.o: In function rpmpd_power_on':
rpmpd.c:(.text+0x3b4): undefined reference to qcom_rpm_smd_write' drivers/soc/qcom/rpmpd.o: In function rpmpd_power_off':
rpmpd.c:(.text+0x520): undefined reference to `qcom_rpm_smd_write'
make: *** [vmlinux] Error 1

Fix it by making QCOM_RPMPD depend on QCOM_SMD_RPM=y

---
Andy, this one should be applied on top of my v11 to add rpmpd/rpmhpd drivers [1] and the fix from Bjorn to drop A family RPM dependency [2]

[1] https://lkml.org/lkml/2019/1/9/1257
[2] https://lkml.org/lkml/2019/1/17/5

 drivers/soc/qcom/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/soc/qcom/Kconfig +1 -1
index a5d5167c3f16..1ee298f6bf17 100644
@@ -109,7 +109,7 @@ config QCOM_RPMHPD
 
 config QCOM_RPMPD
 	bool "Qualcomm RPM Power domain driver"
-	depends on QCOM_SMD_RPM
+	depends on QCOM_SMD_RPM=y
 	help
 	  QCOM RPM Power domain driver to support power-domains with
 	  performance states. The driver communicates a performance state
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
#1 lore raw

Quoting Rajendra Nayak (2019-01-17 20:48:01)

drivers/soc/qcom/Kconfig config QCOM_RPMHPD · line 111

Quoted 11 lines
 drivers/soc/qcom/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/soc/qcom/Kconfig +0 -0
index a5d5167c3f16..1ee298f6bf17 100644
@@ -109,7 +109,7 @@ config QCOM_RPMHPD
 
 config QCOM_RPMPD
        bool "Qualcomm RPM Power domain driver"

Just curious, does it need to be bool for some reason?

#2 lore raw
Quoted 10 lines · Stephen Boyd

Quoting Rajendra Nayak (2019-01-17 20:48:01)

Quoted 9 lines
  drivers/soc/qcom/Kconfig | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
drivers/soc/qcom/Kconfig +0 -0
index a5d5167c3f16..1ee298f6bf17 100644
@@ -109,7 +109,7 @@ config QCOM_RPMHPD
 
Quoted 4 lines

drivers/soc/qcom/Kconfig config QCOM_RPMHPD · line 111

Quoted 2 lines
drivers/soc/qcom/Kconfig +0 -0
  config QCOM_RPMPD
         bool "Qualcomm RPM Power domain driver"

Just curious, does it need to be bool for some reason?

Here's the link to the discussion around it on the v3 patchset of this series
https://lkml.org/lkml/2018/6/12/111

Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
#3 lore raw
Quoted 17 lines · Rajendra Nayak
Quoted 4 lines · Stephen Boyd

Quoting Rajendra Nayak (2019-01-17 20:48:01)

Quoted 3 lines

  drivers/soc/qcom/Kconfig | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
Quoted 9 lines
Quoted 7 lines
drivers/soc/qcom/Kconfig +0 -0
index a5d5167c3f16..1ee298f6bf17 100644
@@ -109,7 +109,7 @@ config QCOM_RPMHPD

  config QCOM_RPMPD
         bool "Qualcomm RPM Power domain driver"

Just curious, does it need to be bool for some reason?

Quoted 3 lines

Here's the link to the discussion around it on the v3 patchset of this series
https://lkml.org/lkml/2018/6/12/111

I think you were missing the implication of "if possible", the implication being that the driver can now not be used in a multi-platform image, and that having a bool driver depend on tristate drivers doesn't make much if any sense.

The argument in the exchange is odd - one can not remove any driver as long there are client drivers / devices attached to it. This is true for all drivers, not just for this one, and handled quite nicely by the driver core. Just run "lsmod" and try to remove a driver with any attached users.

Guenter

#4 lore raw

Adding Viresh and Ulf to this thread,

Quoted 18 lines · Guenter Roeck
Quoted 17 lines · Rajendra Nayak
Quoted 13 lines · Stephen Boyd

Quoting Rajendra Nayak (2019-01-17 20:48:01)

Quoted 10 lines

  drivers/soc/qcom/Kconfig | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
drivers/soc/qcom/Kconfig +0 -0
index a5d5167c3f16..1ee298f6bf17 100644
@@ -109,7 +109,7 @@ config QCOM_RPMHPD

  config QCOM_RPMPD
         bool "Qualcomm RPM Power domain driver"

Just curious, does it need to be bool for some reason?

Quoted 7 lines
Quoted 3 lines

Here's the link to the discussion around it on the v3 patchset of this series
https://lkml.org/lkml/2018/6/12/111

I think you were missing the implication of "if possible", the implication being that the driver can now not be used in a multi-platform image, and that having a bool driver depend on tristate drivers doesn't make much if any sense.

Fair enough, I hadn't thought about the implications with multi platform builds, I guess it makes sense to have these drivers as tristate then.

I was doing some quick testing by adding the calls to of_genpd_remove_last() as suggested by Ulf for cleaning up the genpd registrations, and I run into this backtrace when the driver re-probes followed by a remove

[   59.204525] kobject ((____ptrval____)): tried to init an initialized object, something is seriously wr.
[   59.214262] CPU: 3 PID: 1600 Comm: sh Not tainted 5.0.0-rc1-00012-g0513915837c5-dirty #32
[   59.222500] Hardware name: Qualcomm Technologies, Inc. SDM845 MTP (DT)
[   59.229081] Call trace:
[   59.231574]  dump_backtrace+0x0/0x148
[   59.235276]  show_stack+0x14/0x20
[   59.238631]  dump_stack+0x8c/0xac
[   59.241980]  kobject_init+0x8c/0xa0
[   59.245506]  device_initialize+0x34/0xc8
[   59.249474]  pm_genpd_init+0x170/0x260
[   59.253261]  rpmhpd_probe+0x194/0x2b0
[   59.256966]  platform_drv_probe+0x4c/0xa8
[   59.261011]  really_probe+0x1e4/0x2c8
[   59.264711]  driver_probe_device+0x58/0x10
[   59.268927]  bind_store+0xdc/0x178
[   59.272356]  drv_attr_store+0x20/0x30
[   59.276061]  sysfs_kf_write+0x48/0x58
[   59.279760]  kernfs_fop_write+0xcc/0x1c8
[   59.283728]  __vfs_write+0x34/0x170
[   59.287245]  vfs_write+0xa8/0x1b8
[   59.290592]  ksys_write+0x5c/0xc8
[   59.293941]  __arm64_sys_write+0x14/0x20
[   59.297897]  el0_svc_common+0xb4/0x118
[   59.301676]  el0_svc_handler+0x2c/0x80
[   59.305455]  el0_svc+0x8/0xc

Viresh/Ulf, looks like we need some cleanup of whats done by device_initialize() in pm_genpd_init() to happen as part of the pm_genpd_remove()?

Quoted 7 lines

The argument in the exchange is odd - one can not remove any driver as long there are client drivers / devices attached to it. This is true for all drivers, not just for this one, and handled quite nicely by the driver core. Just run "lsmod" and try to remove a driver with any attached users.

Guenter

Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
#5 lore raw
Quoted 30 lines · Rajendra Nayak

I was doing some quick testing by adding the calls to of_genpd_remove_last() as suggested by Ulf for cleaning up the genpd registrations, and I run into this backtrace when the driver re-probes followed by a remove

[   59.204525] kobject ((____ptrval____)): tried to init an initialized object, something is seriously wr.
[   59.214262] CPU: 3 PID: 1600 Comm: sh Not tainted 5.0.0-rc1-00012-g0513915837c5-dirty #32
[   59.222500] Hardware name: Qualcomm Technologies, Inc. SDM845 MTP (DT)
[   59.229081] Call trace:
[   59.231574]  dump_backtrace+0x0/0x148
[   59.235276]  show_stack+0x14/0x20
[   59.238631]  dump_stack+0x8c/0xac
[   59.241980]  kobject_init+0x8c/0xa0
[   59.245506]  device_initialize+0x34/0xc8
[   59.249474]  pm_genpd_init+0x170/0x260
[   59.253261]  rpmhpd_probe+0x194/0x2b0
[   59.256966]  platform_drv_probe+0x4c/0xa8
[   59.261011]  really_probe+0x1e4/0x2c8
[   59.264711]  driver_probe_device+0x58/0x10
[   59.268927]  bind_store+0xdc/0x178
[   59.272356]  drv_attr_store+0x20/0x30
[   59.276061]  sysfs_kf_write+0x48/0x58
[   59.279760]  kernfs_fop_write+0xcc/0x1c8
[   59.283728]  __vfs_write+0x34/0x170
[   59.287245]  vfs_write+0xa8/0x1b8
[   59.290592]  ksys_write+0x5c/0xc8
[   59.293941]  __arm64_sys_write+0x14/0x20
[   59.297897]  el0_svc_common+0xb4/0x118
[   59.301676]  el0_svc_handler+0x2c/0x80
[   59.305455]  el0_svc+0x8/0xc
Quoted 2 lines

Viresh/Ulf, looks like we need some cleanup of whats done by device_initialize() in pm_genpd_init() to happen as part of the pm_genpd_remove()?

Yeah, we normally need to call put_device() for that. Maybe something like this:

drivers/base/power/domain.c +1 -0
index 500de1dee967..d5b984f042ec 100644
@@ -1812,6 +1812,7 @@ static int genpd_remove(struct generic_pm_domain *genpd)
        genpd_unlock(genpd);
        cancel_work_sync(&genpd->power_off_work);
        kfree(genpd->free);
+       put_device(&genpd->dev);
        pr_debug("%s: removed %s\n", __func__, genpd->name);
 
        return 0;
Signature · 1 line
viresh
#6 lore raw
Quoted 41 lines · Viresh Kumar
Quoted 30 lines · Rajendra Nayak

I was doing some quick testing by adding the calls to of_genpd_remove_last() as suggested by Ulf for cleaning up the genpd registrations, and I run into this backtrace when the driver re-probes followed by a remove

[   59.204525] kobject ((____ptrval____)): tried to init an initialized object, something is seriously wr.
[   59.214262] CPU: 3 PID: 1600 Comm: sh Not tainted 5.0.0-rc1-00012-g0513915837c5-dirty #32
[   59.222500] Hardware name: Qualcomm Technologies, Inc. SDM845 MTP (DT)
[   59.229081] Call trace:
[   59.231574]  dump_backtrace+0x0/0x148
[   59.235276]  show_stack+0x14/0x20
[   59.238631]  dump_stack+0x8c/0xac
[   59.241980]  kobject_init+0x8c/0xa0
[   59.245506]  device_initialize+0x34/0xc8
[   59.249474]  pm_genpd_init+0x170/0x260
[   59.253261]  rpmhpd_probe+0x194/0x2b0
[   59.256966]  platform_drv_probe+0x4c/0xa8
[   59.261011]  really_probe+0x1e4/0x2c8
[   59.264711]  driver_probe_device+0x58/0x10
[   59.268927]  bind_store+0xdc/0x178
[   59.272356]  drv_attr_store+0x20/0x30
[   59.276061]  sysfs_kf_write+0x48/0x58
[   59.279760]  kernfs_fop_write+0xcc/0x1c8
[   59.283728]  __vfs_write+0x34/0x170
[   59.287245]  vfs_write+0xa8/0x1b8
[   59.290592]  ksys_write+0x5c/0xc8
[   59.293941]  __arm64_sys_write+0x14/0x20
[   59.297897]  el0_svc_common+0xb4/0x118
[   59.301676]  el0_svc_handler+0x2c/0x80
[   59.305455]  el0_svc+0x8/0xc
Quoted 2 lines

Viresh/Ulf, looks like we need some cleanup of whats done by device_initialize() in pm_genpd_init() to happen as part of the pm_genpd_remove()?

Yeah, we normally need to call put_device() for that. Maybe something like this:

drivers/base/power/domain.c +0 -0
index 500de1dee967..d5b984f042ec 100644

drivers/base/power/domain.c static int genpd_remove(struct generic_pm_domain *genpd) · line 1818

Quoted 7 lines
drivers/base/power/domain.c +1 -0
@@ -1812,6 +1812,7 @@ static int genpd_remove(struct generic_pm_domain *genpd)
         genpd_unlock(genpd);
         cancel_work_sync(&genpd->power_off_work);
         kfree(genpd->free);
+       put_device(&genpd->dev);
         pr_debug("%s: removed %s\n", __func__, genpd->name);
 
         return 0;

Seeing this now during remove

[   34.008216] ------------[ cut here ]------------
[   34.012894] Device 'mss' does not have a release() function, it is broken and must be fixed. See Docum.
[   34.024143] WARNING: CPU: 6 PID: 1600 at drivers/base/core.c:922 device_release+0x84/0x98
[   34.032379] Modules linked in:
[   34.035477] CPU: 6 PID: 1600 Comm: sh Not tainted 5.0.0-rc1-00012-g0513915837c5-dirty #34
[   34.043724] Hardware name: Qualcomm Technologies, Inc. SDM845 MTP (DT)
[   34.050305] pstate: 60400005 (nZCv daif +PAN -UAO)
[   34.055141] pc : device_release+0x84/0x98
[   34.059186] lr : device_release+0x84/0x98
[   34.063229] sp : ffff0000183c3b40
[   34.066574] x29: ffff0000183c3b40 x28: ffff800174b55100
[   34.071930] x27: 0000000000000000 x26: 0000000000000000
[   34.077285] x25: 0000000056000000 x24: 0000000000000015
[   34.082640] x23: dead000000000100 x22: dead000000000200
[   34.087997] x21: 0000000000000000 x20: ffff0000112f9780
[   34.093352] x19: ffff0000112f9770 x18: ffff0000111dd6c8
[   34.098719] x17: 0000000000000000 x16: 0000000000000000
[   34.104072] x15: ffff0000983c3867 x14: 6966206562207473
[   34.109438] x13: 756d20646e61206e x12: 656b6f7262207369
[   34.114791] x11: 207469202c6e6f69 x10: 74636e7566202928
[   34.120147] x9 : 657361656c657220 x8 : 742e7463656a626f
[   34.125514] x7 : 6b2f6e6f69746174 x6 : 0000000000000149
[   34.130867] x5 : 0000000000000020 x4 : 000000000001c200
[   34.136223] x3 : 00000000ffffffff x2 : ffff0000111f5828
[   34.141579] x1 : bc1d8ed5415ab200 x0 : 0000000000000000
[   34.146936] Call trace:
[   34.149413]  device_release+0x84/0x98
[   34.153116]  kobject_put+0x98/0xe8
[   34.156551]  put_device+0x14/0x28
[   34.159899]  genpd_remove+0x108/0x190
[   34.163598]  of_genpd_remove_last+0xbc/0xf0
[   34.167828]  rpmhpd_remove+0x2c/0x48
[   34.171436]  platform_drv_remove+0x24/0x68
[   34.175574]  device_release_driver_internal+0x184/0x218
[   34.180845]  device_release_driver+0x14/0x20
[   34.185152]  unbind_store+0xec/0x138
[   34.188760]  drv_attr_store+0x20/0x30
[   34.192465]  sysfs_kf_write+0x48/0x58
[   34.196163]  kernfs_fop_write+0xcc/0x1c8
[   34.200129]  __vfs_write+0x34/0x170
[   34.203654]  vfs_write+0xa8/0x1b8
[   34.206998]  ksys_write+0x5c/0xc8
[   34.210345]  __arm64_sys_write+0x14/0x20
[   34.214318]  el0_svc_common+0xb4/0x118
[   34.218107]  el0_svc_handler+0x2c/0x80
[   34.221899]  el0_svc+0x8/0xc
[   34.224804] ---[ end trace 9cee1458d0b6c175 ]---
[   34.229490] ------------[ cut here ]------------
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
#7 lore raw
Quoted 42 lines · Rajendra Nayak

Seeing this now during remove

[   34.008216] ------------[ cut here ]------------
[   34.012894] Device 'mss' does not have a release() function, it is broken and must be fixed. See Docum.
[   34.024143] WARNING: CPU: 6 PID: 1600 at drivers/base/core.c:922 device_release+0x84/0x98
[   34.032379] Modules linked in:
[   34.035477] CPU: 6 PID: 1600 Comm: sh Not tainted 5.0.0-rc1-00012-g0513915837c5-dirty #34
[   34.043724] Hardware name: Qualcomm Technologies, Inc. SDM845 MTP (DT)
[   34.050305] pstate: 60400005 (nZCv daif +PAN -UAO)
[   34.055141] pc : device_release+0x84/0x98
[   34.059186] lr : device_release+0x84/0x98
[   34.063229] sp : ffff0000183c3b40
[   34.066574] x29: ffff0000183c3b40 x28: ffff800174b55100
[   34.071930] x27: 0000000000000000 x26: 0000000000000000
[   34.077285] x25: 0000000056000000 x24: 0000000000000015
[   34.082640] x23: dead000000000100 x22: dead000000000200
[   34.087997] x21: 0000000000000000 x20: ffff0000112f9780
[   34.093352] x19: ffff0000112f9770 x18: ffff0000111dd6c8
[   34.098719] x17: 0000000000000000 x16: 0000000000000000
[   34.104072] x15: ffff0000983c3867 x14: 6966206562207473
[   34.109438] x13: 756d20646e61206e x12: 656b6f7262207369
[   34.114791] x11: 207469202c6e6f69 x10: 74636e7566202928
[   34.120147] x9 : 657361656c657220 x8 : 742e7463656a626f
[   34.125514] x7 : 6b2f6e6f69746174 x6 : 0000000000000149
[   34.130867] x5 : 0000000000000020 x4 : 000000000001c200
[   34.136223] x3 : 00000000ffffffff x2 : ffff0000111f5828
[   34.141579] x1 : bc1d8ed5415ab200 x0 : 0000000000000000
[   34.146936] Call trace:
[   34.149413]  device_release+0x84/0x98
[   34.153116]  kobject_put+0x98/0xe8
[   34.156551]  put_device+0x14/0x28
[   34.159899]  genpd_remove+0x108/0x190
[   34.163598]  of_genpd_remove_last+0xbc/0xf0
[   34.167828]  rpmhpd_remove+0x2c/0x48
[   34.171436]  platform_drv_remove+0x24/0x68
[   34.175574]  device_release_driver_internal+0x184/0x218
[   34.180845]  device_release_driver+0x14/0x20
[   34.185152]  unbind_store+0xec/0x138
[   34.188760]  drv_attr_store+0x20/0x30
[   34.192465]  sysfs_kf_write+0x48/0x58
[   34.196163]  kernfs_fop_write+0xcc/0x1c8
[   34.200129]  __vfs_write+0x34/0x170
Quoted 8 lines
[   34.203654]  vfs_write+0xa8/0x1b8
[   34.206998]  ksys_write+0x5c/0xc8
[   34.210345]  __arm64_sys_write+0x14/0x20
[   34.214318]  el0_svc_common+0xb4/0x118
[   34.218107]  el0_svc_handler+0x2c/0x80
[   34.221899]  el0_svc+0x8/0xc
[   34.224804] ---[ end trace 9cee1458d0b6c175 ]---
[   34.229490] ------------[ cut here ]------------

This shall fix it:

drivers/base/power/domain.c +10 -0
index 500de1dee967..8a44b1f356d0 100644
@@ -1711,6 +1711,15 @@ static void genpd_lock_init(struct generic_pm_domain *genpd)
        }
 }
 
+static void genpd_device_release(struct device *dev)
+{
+}
+
+struct device_type genpd_device_type = {
+       .name =         "genpd_device",
+       .release =      genpd_device_release,
+};
+
 /**
  * pm_genpd_init - Initialize a generic I/O PM domain object.
  * @genpd: PM domain object to initialize.
@@ -1770,6 +1779,7 @@ int pm_genpd_init(struct generic_pm_domain *genpd,
                pr_warn("%s : no governor for states\n", genpd->name);
        }
 
+       genpd->dev.type = &genpd_device_type;
        device_initialize(&genpd->dev);
        dev_set_name(&genpd->dev, "%s", genpd->name);
 

@Ulf: You fine with this ?

Signature · 1 line
viresh
#8 lore raw
Quoted 59 lines · Viresh Kumar
Quoted 42 lines · Rajendra Nayak

Seeing this now during remove

[   34.008216] ------------[ cut here ]------------
[   34.012894] Device 'mss' does not have a release() function, it is broken and must be fixed. See Docum.
[   34.024143] WARNING: CPU: 6 PID: 1600 at drivers/base/core.c:922 device_release+0x84/0x98
[   34.032379] Modules linked in:
[   34.035477] CPU: 6 PID: 1600 Comm: sh Not tainted 5.0.0-rc1-00012-g0513915837c5-dirty #34
[   34.043724] Hardware name: Qualcomm Technologies, Inc. SDM845 MTP (DT)
[   34.050305] pstate: 60400005 (nZCv daif +PAN -UAO)
[   34.055141] pc : device_release+0x84/0x98
[   34.059186] lr : device_release+0x84/0x98
[   34.063229] sp : ffff0000183c3b40
[   34.066574] x29: ffff0000183c3b40 x28: ffff800174b55100
[   34.071930] x27: 0000000000000000 x26: 0000000000000000
[   34.077285] x25: 0000000056000000 x24: 0000000000000015
[   34.082640] x23: dead000000000100 x22: dead000000000200
[   34.087997] x21: 0000000000000000 x20: ffff0000112f9780
[   34.093352] x19: ffff0000112f9770 x18: ffff0000111dd6c8
[   34.098719] x17: 0000000000000000 x16: 0000000000000000
[   34.104072] x15: ffff0000983c3867 x14: 6966206562207473
[   34.109438] x13: 756d20646e61206e x12: 656b6f7262207369
[   34.114791] x11: 207469202c6e6f69 x10: 74636e7566202928
[   34.120147] x9 : 657361656c657220 x8 : 742e7463656a626f
[   34.125514] x7 : 6b2f6e6f69746174 x6 : 0000000000000149
[   34.130867] x5 : 0000000000000020 x4 : 000000000001c200
[   34.136223] x3 : 00000000ffffffff x2 : ffff0000111f5828
[   34.141579] x1 : bc1d8ed5415ab200 x0 : 0000000000000000
[   34.146936] Call trace:
[   34.149413]  device_release+0x84/0x98
[   34.153116]  kobject_put+0x98/0xe8
[   34.156551]  put_device+0x14/0x28
[   34.159899]  genpd_remove+0x108/0x190
[   34.163598]  of_genpd_remove_last+0xbc/0xf0
[   34.167828]  rpmhpd_remove+0x2c/0x48
[   34.171436]  platform_drv_remove+0x24/0x68
[   34.175574]  device_release_driver_internal+0x184/0x218
[   34.180845]  device_release_driver+0x14/0x20
[   34.185152]  unbind_store+0xec/0x138
[   34.188760]  drv_attr_store+0x20/0x30
[   34.192465]  sysfs_kf_write+0x48/0x58
[   34.196163]  kernfs_fop_write+0xcc/0x1c8
[   34.200129]  __vfs_write+0x34/0x170
Quoted 8 lines
[   34.203654]  vfs_write+0xa8/0x1b8
[   34.206998]  ksys_write+0x5c/0xc8
[   34.210345]  __arm64_sys_write+0x14/0x20
[   34.214318]  el0_svc_common+0xb4/0x118
[   34.218107]  el0_svc_handler+0x2c/0x80
[   34.221899]  el0_svc+0x8/0xc
[   34.224804] ---[ end trace 9cee1458d0b6c175 ]---
[   34.229490] ------------[ cut here ]------------

This shall fix it:

drivers/base/power/domain.c +0 -0
index 500de1dee967..8a44b1f356d0 100644

drivers/base/power/domain.c static void genpd_lock_init(struct generic_pm_domain *genpd) · line 1717

Quoted 7 lines
drivers/base/power/domain.c +4 -0
@@ -1711,6 +1711,15 @@ static void genpd_lock_init(struct generic_pm_domain *genpd)
         }
  }
 
+static void genpd_device_release(struct device *dev)
+{
+}
+

See Documentation/kobject.txt.

"One important point cannot be overstated: every kobject must have a release() method, and the kobject must persist (in a consistent state) until that method is called. If these constraints are not met, the code is flawed. Note that the kernel will warn you if you forget to provide a release() method. Do not try to get rid of this warning by providing an "empty" release function.

If all your cleanup function needs to do is call kfree(), then you must create a wrapper function which uses container_of() to upcast to the correct type (as shown in the example above) and then calls kfree() on the overall structure."

Guenter

Quoted 19 lines
drivers/base/power/domain.c +6 -0
+struct device_type genpd_device_type = {
+       .name =         "genpd_device",
+       .release =      genpd_device_release,
+};
+
  /**
   * pm_genpd_init - Initialize a generic I/O PM domain object.
   * @genpd: PM domain object to initialize.
@@ -1770,6 +1779,7 @@ int pm_genpd_init(struct generic_pm_domain *genpd,
                 pr_warn("%s : no governor for states\n", genpd->name);
         }
 
+       genpd->dev.type = &genpd_device_type;
         device_initialize(&genpd->dev);
         dev_set_name(&genpd->dev, "%s", genpd->name);
 

@Ulf: You fine with this ?

#9 lore raw
Quoted 9 lines · Guenter Roeck

See Documentation/kobject.txt.

"One important point cannot be overstated: every kobject must have a release() method, and the kobject must persist (in a consistent state) until that method is called. If these constraints are not met, the code is flawed. Note that the kernel will warn you if you forget to provide a release() method. Do not try to get rid of this warning by providing an "empty" release function.

Quoted 4 lines

If all your cleanup function needs to do is call kfree(), then you must create a wrapper function which uses container_of() to upcast to the correct type (as shown in the example above) and then calls kfree() on the overall structure."

Guenter, you are of course correct but I am not sure what else we can do here. In the most common case device & kobject are allocated along with the real entity, like struct genpd here, and we free those from the release routine (i.e. kfree(genpd)). But in our case most of the times the genpd is created statically by platform drivers and not by the genpd core, and so we can't free it from release().

Over that the genpd->dev is only used by the OPP core currently and the device isn't even registered with the device core.

Signature · 1 line
viresh
#10 lore raw
In reply to Stephen Boyd
Quoted 10 lines · Stephen Boyd

Quoting Rajendra Nayak (2019-01-17 20:48:01)

Quoted 9 lines
 drivers/soc/qcom/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/soc/qcom/Kconfig +0 -0
index a5d5167c3f16..1ee298f6bf17 100644
@@ -109,7 +109,7 @@ config QCOM_RPMHPD
 
Quoted 5 lines

drivers/soc/qcom/Kconfig config QCOM_RPMHPD · line 111

Quoted 2 lines
drivers/soc/qcom/Kconfig +0 -0
 config QCOM_RPMPD
        bool "Qualcomm RPM Power domain driver"

Just curious, does it need to be bool for some reason?

It's unfortunately not possible to have any genpd, iommu or pinctrl drivers compiled as modules, because once you pass lateinit probe deferral is purposefully broken. See driver_deferred_probe_check_state().

This also means that if you're unlucky and your kernel reached lateinit before the SMD communication with RPM is established and has brought up the rpmpd, you are left with a completely broken system. Unfortunately this isn't that hard to reproduce with a minimal defconfig.

Regards,
Bjorn