introduce support for early platform drivers

From: Bartosz Golaszewski <[email protected]>

Add a new single-bit field to struct device which will be used to indicate that a device was probed early in the boot sequence.

This does not affect the size of struct device on any architecture I built on (ARM, ARM64, x86_64).

Signed-off-by Bartosz Golaszewski <[email protected]>

From: Bartosz Golaszewski <[email protected]>

Early platform devices can have devres objects allocated in early_probe(). This is not a bug so don't dump stack in this case.

Signed-off-by Bartosz Golaszewski <[email protected]>

From: Bartosz Golaszewski <[email protected]>

We will use this function to match devices in the early platform core code. Export it locally in drivers/base.

Signed-off-by Bartosz Golaszewski <[email protected]>

From: Bartosz Golaszewski <[email protected]>

The early platform driver framework will allocate platform devices using its own internal mechanisms. Provide a function that allows to initialize a platform device object without allocating it.

Signed-off-by Bartosz Golaszewski <[email protected]>

From: Bartosz Golaszewski <[email protected]>

The early platform driver mechanism will use this function as the device release callback. Make it available in drivers/base.

Signed-off-by Bartosz Golaszewski <[email protected]>

From: Bartosz Golaszewski <[email protected]>

This flag indicates that a device node has been populated early and that it needs different handling when called from of_platform_populate().

Signed-off-by Bartosz Golaszewski <[email protected]>

From: Bartosz Golaszewski <[email protected]>

The early platform driver framework will need to setup the device resources before the regular populating of the device tree happens.

Provide a separate function for that.

Signed-off-by Bartosz Golaszewski <[email protected]>

From: Bartosz Golaszewski <[email protected]>

The early platform device framework will need to initialize the platform device objects without them being allocated in of_device_alloc(). Provide a routine that allows it.

Signed-off-by Bartosz Golaszewski <[email protected]>

From: Bartosz Golaszewski <[email protected]>

Provide a separate section in which pointers to early platform driver structs will be stored.

Signed-off-by Bartosz Golaszewski <[email protected]>

Hi Bartosz,

Quoted 6 lines · Bartosz Golaszewski

From: Bartosz Golaszewski <[email protected]>

Provide a separate section in which pointers to early platform driver structs will be stored.

Signed-off-by Bartosz Golaszewski <[email protected]>

Thanks for your patch!

Quoted 6 lines

include/asm-generic/vmlinux.lds.h +0 -0

@@ -214,6 +214,16 @@
 #define CPU_METHOD_OF_TABLES() OF_TABLE(CONFIG_SMP, cpu_method)
 #define CPUIDLE_METHOD_OF_TABLES() OF_TABLE(CONFIG_CPU_IDLE, cpuidle_method)
 

include/asm-generic/vmlinux.lds.h line 219

Quoted 3 lines

include/asm-generic/vmlinux.lds.h +3 -0

+#ifdef CONFIG_EARLY_PLATFORM
+#define EARLY_PLATFORM_DRIVERS_TABLE()                                 \
+       . = ALIGN(8);                                                   \

Should this use STRUCT_ALIGN() instead?

include/asm-generic/vmlinux.lds.h line 225

Quoted 6 lines

include/asm-generic/vmlinux.lds.h +6 -0

+       VMLINUX_SYMBOL(__early_platform_drivers_table) = .;             \
+       KEEP(*(__early_platform_drivers_table))                         \
+       VMLINUX_SYMBOL(__early_platform_drivers_table_end) = .;
+#else
+#define EARLY_PLATFORM_DRIVERS_TABLE()
+#endif

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

From: Bartosz Golaszewski <[email protected]>

This introduces the core part of support for early platform drivers and devices.

+ int (*early_probe)(struct platform_device *);

Quoted 5 lines

From: Bartosz Golaszewski <[email protected]>

This introduces the core part of support for early platform drivers and devices.

It looks like most of your prep patches are to separate the alloc and init of platform devices because you are essentially making early devices/drivers a sub-class. Maybe you could avoid doing that and simplify things a bit. Comments below based on doing that...

Quoted 5 lines

+/**
+ * struct early_platform_driver
+ *
+ * @pdrv: real platform driver associated with this early platform driver
+ * @list: list head for the list of early platform drivers

include/linux/early_platform.h line 24

Quoted 5 lines

+ * @early_probe: early probe callback
+ */

include/linux/early_platform.h new file +3 -0

+struct early_platform_driver {
+       struct platform_driver pdrv;
+       struct list_head list;

Couldn't you use an existing list in driver_private until you move over to the normal bus infra.

Quoted 1 line

+ int (*early_probe)(struct platform_device *);

Just add this to platform_driver.

include/linux/early_platform.h line 38

Quoted 13 lines

include/linux/early_platform.h new file +13 -0

+};
+
+/**
+ * struct early_platform_device
+ *
+ * @pdev: real platform device associated with this early platform device
+ * @list: list head for the list of early platform devices
+ * @deferred: true if this device's early probe was deferred
+ * @deferred_drv: early platform driver with which this device was matched
+ */
+struct early_platform_device {
+       struct platform_device pdev;
+       struct list_head list;

Use a list in device_private?

include/linux/early_platform.h line 40

Quoted 2 lines

include/linux/early_platform.h new file +2 -0

+       bool deferred;
+       struct early_platform_driver *deferred_drv;

Can't you use the existing deferred probe list?

2018-05-14 15:37 GMT+02:00 Rob Herring <[email protected]>:

Quoted 12 lines · Bartosz Golaszewski

Quoted 5 lines

From: Bartosz Golaszewski <[email protected]>

This introduces the core part of support for early platform drivers and devices.

It looks like most of your prep patches are to separate the alloc and init of platform devices because you are essentially making early devices/drivers a sub-class. Maybe you could avoid doing that and simplify things a bit. Comments below based on doing that...

My aim was to change as little as possible for everybody else while fixing our problem. These changes are already controversial enough without risky reusing of existing fields in common structures. I was just afraid that there are too many intricacies for it to be safe.

Quoted 5 lines

Quoted 5 lines

+/**
+ * struct early_platform_driver
+ *
+ * @pdrv: real platform driver associated with this early platform driver
+ * @list: list head for the list of early platform drivers

Quoted 9 lines

include/linux/early_platform.h line 24

Quoted 5 lines

+ * @early_probe: early probe callback
+ */

include/linux/early_platform.h new file +3 -0

+struct early_platform_driver {
+       struct platform_driver pdrv;
+       struct list_head list;

Couldn't you use an existing list in driver_private until you move over to the normal bus infra.

This is something that the previous implementation did. It was quite unreadable, so I decided to go with a separate list.

Quoted 4 lines

Quoted 1 line

+ int (*early_probe)(struct platform_device *);

Just add this to platform_driver.

This would extend the structure for everybody else while there'll be very few such devices and not all systems would even require it.

Quoted 19 lines

include/linux/early_platform.h line 38

Quoted 13 lines

include/linux/early_platform.h new file +13 -0

+};
+
+/**
+ * struct early_platform_device
+ *
+ * @pdev: real platform device associated with this early platform device
+ * @list: list head for the list of early platform devices
+ * @deferred: true if this device's early probe was deferred
+ * @deferred_drv: early platform driver with which this device was matched
+ */
+struct early_platform_device {
+       struct platform_device pdev;
+       struct list_head list;

Use a list in device_private?

Quoted 2 lines

include/linux/early_platform.h new file +2 -0

+       bool deferred;
+       struct early_platform_driver *deferred_drv;

Quoted 2 lines

Can't you use the existing deferred probe list?

I thought about it, but I was afraid there could be some timing issues with that and decided against it. The early deferral also doesn't work in a workque, but is synchronous instead.

From: Bartosz Golaszewski <[email protected]>

Check the relevant flag in the device node and skip the allocation part for devices that were populated early.

Signed-off-by Bartosz Golaszewski <[email protected]>

drivers/of/platform.c +1 -3

@@ -196,9 +197,17 @@ static struct platform_device *of_platform_device_create_pdata(
            of_node_test_and_set_flag(np, OF_POPULATED))
                return NULL;
 
-       dev = of_device_alloc(np, bus_id, parent);
-       if (!dev)
-               goto err_clear_flag;
+       if (of_node_check_flag(np, OF_POPULATED_EARLY)) {

This series is a follow-up to the RFC[1] posted a couple days ago.

NOTE: this series applies on top of my recent patches[2] that move the previous implementation of early platform devices to arch/sh.

Problem:

Certain class of devices, such as timers, certain clock drivers and irq chip drivers need to be probed early in the boot sequence. The currently preferred approach is using one of the OF_DECLARE() macros. This however does not create a platform device which has many drawbacks - such as not being able to use devres routines, dev_ log functions or no way of deferring the init OF function if some other resources are missing.

For drivers that use both platform drivers and OF_DECLARE the situation is even more complicated as the code needs to take into account that there can possibly be no struct device present. For a specific use case that we're having problems with, please refer to the recent DaVinci common-clock conversion patches and the nasty workaround that this problem implies[3].

We also used to have an early platform drivers implementation but they were not integrated with the linux device model at all - they merely used the same data structures. The users could not use devres, defer probe and the early devices never became actual platform devices later on.

Proposed solution:

This series aims at solving this problem by (re-)introducing the concept of early platform drivers and devices - this time however in a way that seamlessly integrates with the existing platform drivers and also offers device-tree support.

The idea is to provide a way for users to probe devices early, while already being able to use devres, devices resources and properties and also deferred probing.

New structures are introduced: the early platform driver contains the early_probe callback which has the same signature as regular platform_device probe. This callback is called early on. The user can have both the early and regular probe speficied or only one of them and they both receive the same platform device object as argument. Any device data allocated early will be carried over to the normal probe.

The architecture code is responsible for calling early_platform_start() in which the early drivers will be registered and devices populated from DT.

Once the device and kobject mechanisms are ready, all early drivers and devices will be converted into real platform drivers and devices. Also: if any of the early platform registration functions will be called once early initialization is done, these functions will work like regular platform_device/driver ones.

Quoted 5 lines

This series is a follow-up to the RFC[1] posted a couple days ago.

NOTE: this series applies on top of my recent patches[2] that move the previous implementation of early platform devices to arch/sh.

Quoted 8 lines

Problem:

Certain class of devices, such as timers, certain clock drivers and irq chip drivers need to be probed early in the boot sequence. The currently preferred approach is using one of the OF_DECLARE() macros. This however does not create a platform device which has many drawbacks - such as not being able to use devres routines, dev_ log functions or no way of deferring the init OF function if some other resources are missing.

I skimmed though this and it doesn't look horrible (how's that for positive feedback? ;) ). But before going into the details, I think first there needs to be agreement this is the right direction.

The question does remain though as to whether this class of devices should be platform drivers. They can't be modules. They can't be hotplugged. Can they be runtime-pm enabled? So the advantage is ...

I assume that the clock maintainers had some reason to move clocks to be platform drivers. It's just not clear to me what that was.

Quoted 5 lines

For drivers that use both platform drivers and OF_DECLARE the situation is even more complicated as the code needs to take into account that there can possibly be no struct device present. For a specific use case that we're having problems with, please refer to the recent DaVinci common-clock conversion patches and the nasty workaround that this problem implies[3].

So devm_kzalloc will work with this solution? Why did we need devm_kzalloc in the first place? The clocks can never be removed and cleaning up on error paths is kind of pointless. The system would be hosed, right?

Quoted 23 lines

We also used to have an early platform drivers implementation but they were not integrated with the linux device model at all - they merely used the same data structures. The users could not use devres, defer probe and the early devices never became actual platform devices later on.

Proposed solution:

This series aims at solving this problem by (re-)introducing the concept of early platform drivers and devices - this time however in a way that seamlessly integrates with the existing platform drivers and also offers device-tree support.

The idea is to provide a way for users to probe devices early, while already being able to use devres, devices resources and properties and also deferred probing.

New structures are introduced: the early platform driver contains the early_probe callback which has the same signature as regular platform_device probe. This callback is called early on. The user can have both the early and regular probe speficied or only one of them and they both receive the same platform device object as argument. Any device data allocated early will be carried over to the normal probe.

Quoted 2 lines

The architecture code is responsible for calling early_platform_start() in which the early drivers will be registered and devices populated from DT.

Can we really do this in one spot for different devices (clk, timers, irq). The sequence is all very carefully crafted. Platform specific hooks is another thing to consider.

Quoted 4 lines

Once the device and kobject mechanisms are ready, all early drivers and devices will be converted into real platform drivers and devices. Also: if any of the early platform registration functions will be called once early initialization is done, these functions will work like regular platform_device/driver ones.

This could leave devices in a weird state. They've successfully probed early, but then are on the deferred list for normal probe for example.

2018-05-11 22:13 GMT+02:00 Rob Herring <[email protected]>:

Quoted 15 lines · Bartosz Golaszewski

Quoted 5 lines

This series is a follow-up to the RFC[1] posted a couple days ago.

NOTE: this series applies on top of my recent patches[2] that move the previous implementation of early platform devices to arch/sh.

Quoted 8 lines

Problem:

Certain class of devices, such as timers, certain clock drivers and irq chip drivers need to be probed early in the boot sequence. The currently preferred approach is using one of the OF_DECLARE() macros. This however does not create a platform device which has many drawbacks - such as not being able to use devres routines, dev_ log functions or no way of deferring the init OF function if some other resources are missing.

Quoted 8 lines

I skimmed though this and it doesn't look horrible (how's that for positive feedback? ;) ). But before going into the details, I think first there needs to be agreement this is the right direction.

The question does remain though as to whether this class of devices should be platform drivers. They can't be modules. They can't be hotplugged. Can they be runtime-pm enabled? So the advantage is ...

The main (but not the only) advantage for drivers that can both be platform drivers and OF_DECLARE drivers is that we get a single entry point and can reuse code without resorting to checking if (!dev). It results in more consistent code base. Another big advantage is consolidation of device tree and machine code for SoC drivers used in different boards of which some are still using board files and others are defined in DT (see: DaVinci).

Quoted 9 lines

I assume that the clock maintainers had some reason to move clocks to be platform drivers. It's just not clear to me what that was.

Quoted 5 lines

For drivers that use both platform drivers and OF_DECLARE the situation is even more complicated as the code needs to take into account that there can possibly be no struct device present. For a specific use case that we're having problems with, please refer to the recent DaVinci common-clock conversion patches and the nasty workaround that this problem implies[3].

Quoted 5 lines

So devm_kzalloc will work with this solution? Why did we need devm_kzalloc in the first place? The clocks can never be removed and cleaning up on error paths is kind of pointless. The system would be hosed, right?

It depends - not all clocks are necessary for system to boot.

Quoted 23 lines

Quoted 23 lines

We also used to have an early platform drivers implementation but they were not integrated with the linux device model at all - they merely used the same data structures. The users could not use devres, defer probe and the early devices never became actual platform devices later on.

Proposed solution:

This series aims at solving this problem by (re-)introducing the concept of early platform drivers and devices - this time however in a way that seamlessly integrates with the existing platform drivers and also offers device-tree support.

The idea is to provide a way for users to probe devices early, while already being able to use devres, devices resources and properties and also deferred probing.

New structures are introduced: the early platform driver contains the early_probe callback which has the same signature as regular platform_device probe. This callback is called early on. The user can have both the early and regular probe speficied or only one of them and they both receive the same platform device object as argument. Any device data allocated early will be carried over to the normal probe.

Quoted 7 lines

Quoted 2 lines

The architecture code is responsible for calling early_platform_start() in which the early drivers will be registered and devices populated from DT.

Can we really do this in one spot for different devices (clk, timers, irq). The sequence is all very carefully crafted. Platform specific hooks is another thing to consider.

This is why I added support for early probe deferral - so that we can stop caring for the order as long as the drivers are aware of other resources they need and we call early_platform_start() the moment the earliest of the early devices is needed.

Quoted 48 lines

2018-05-11 22:13 GMT+02:00 Rob Herring <[email protected]>:

Quoted 15 lines · Bartosz Golaszewski

Quoted 5 lines

This series is a follow-up to the RFC[1] posted a couple days ago.

NOTE: this series applies on top of my recent patches[2] that move the previous implementation of early platform devices to arch/sh.

Quoted 8 lines

Problem:

Certain class of devices, such as timers, certain clock drivers and irq chip drivers need to be probed early in the boot sequence. The currently preferred approach is using one of the OF_DECLARE() macros. This however does not create a platform device which has many drawbacks - such as not being able to use devres routines, dev_ log functions or no way of deferring the init OF function if some other resources are missing.

Quoted 8 lines

I skimmed though this and it doesn't look horrible (how's that for positive feedback? ;) ). But before going into the details, I think first there needs to be agreement this is the right direction.

The question does remain though as to whether this class of devices should be platform drivers. They can't be modules. They can't be hotplugged. Can they be runtime-pm enabled? So the advantage is ...

The main (but not the only) advantage for drivers that can both be platform drivers and OF_DECLARE drivers is that we get a single entry point and can reuse code without resorting to checking if (!dev). It results in more consistent code base. Another big advantage is consolidation of device tree and machine code for SoC drivers used in different boards of which some are still using board files and others are defined in DT (see: DaVinci).

Quoted 14 lines

I assume that the clock maintainers had some reason to move clocks to be platform drivers. It's just not clear to me what that was.

Quoted 5 lines

For drivers that use both platform drivers and OF_DECLARE the situation is even more complicated as the code needs to take into account that there can possibly be no struct device present. For a specific use case that we're having problems with, please refer to the recent DaVinci common-clock conversion patches and the nasty workaround that this problem implies[3].

So devm_kzalloc will work with this solution? Why did we need devm_kzalloc in the first place? The clocks can never be removed and cleaning up on error paths is kind of pointless. The system would be hosed, right?

Quoted 1 line

It depends - not all clocks are necessary for system to boot.

That doesn't matter. You have a single driver for all/most of the clocks, so the driver can't be removed.

Quoted 32 lines

Quoted 24 lines

Quoted 24 lines

We also used to have an early platform drivers implementation but they were not integrated with the linux device model at all - they merely used the same data structures. The users could not use devres, defer probe and the early devices never became actual platform devices later on.

Proposed solution:

This series aims at solving this problem by (re-)introducing the concept of early platform drivers and devices - this time however in a way that seamlessly integrates with the existing platform drivers and also offers device-tree support.

The idea is to provide a way for users to probe devices early, while already being able to use devres, devices resources and properties and also deferred probing.

New structures are introduced: the early platform driver contains the early_probe callback which has the same signature as regular platform_device probe. This callback is called early on. The user can have both the early and regular probe speficied or only one of them and they both receive the same platform device object as argument. Any device data allocated early will be carried over to the normal probe.

Quoted 7 lines

Quoted 2 lines

The architecture code is responsible for calling early_platform_start() in which the early drivers will be registered and devices populated from DT.

Can we really do this in one spot for different devices (clk, timers, irq). The sequence is all very carefully crafted. Platform specific hooks is another thing to consider.

Quoted 4 lines

This is why I added support for early probe deferral - so that we can stop caring for the order as long as the drivers are aware of other resources they need and we call early_platform_start() the moment the earliest of the early devices is needed.

Deferred probe helps for inter-device dependencies, but I am more concerned about timing of trying to register clocksources, irqchips, etc. What happens if we probe drivers before the infrastructure is initialized?

Rob

Hi Rob,

Quoting Rob Herring (2018-05-14 06:20:57)

Quoted 51 lines · Bartosz Golaszewski

Quoted 49 lines

2018-05-11 22:13 GMT+02:00 Rob Herring <[email protected]>:

Quoted 15 lines · Bartosz Golaszewski

Quoted 5 lines

This series is a follow-up to the RFC[1] posted a couple days ago.

NOTE: this series applies on top of my recent patches[2] that move the previous implementation of early platform devices to arch/sh.

Quoted 8 lines

Problem:

Certain class of devices, such as timers, certain clock drivers and irq chip drivers need to be probed early in the boot sequence. The currently preferred approach is using one of the OF_DECLARE() macros. This however does not create a platform device which has many drawbacks - such as not being able to use devres routines, dev_ log functions or no way of deferring the init OF function if some other resources are missing.

Quoted 8 lines

I skimmed though this and it doesn't look horrible (how's that for positive feedback? ;) ). But before going into the details, I think first there needs to be agreement this is the right direction.

The question does remain though as to whether this class of devices should be platform drivers. They can't be modules. They can't be hotplugged. Can they be runtime-pm enabled? So the advantage is ...

The main (but not the only) advantage for drivers that can both be platform drivers and OF_DECLARE drivers is that we get a single entry point and can reuse code without resorting to checking if (!dev). It results in more consistent code base. Another big advantage is consolidation of device tree and machine code for SoC drivers used in different boards of which some are still using board files and others are defined in DT (see: DaVinci).

Quoted 9 lines

I assume that the clock maintainers had some reason to move clocks to be platform drivers. It's just not clear to me what that was.

Quoted 5 lines

For drivers that use both platform drivers and OF_DECLARE the situation is even more complicated as the code needs to take into account that there can possibly be no struct device present. For a specific use case that we're having problems with, please refer to the recent DaVinci common-clock conversion patches and the nasty workaround that this problem implies[3].

Quoted 5 lines

So devm_kzalloc will work with this solution? Why did we need devm_kzalloc in the first place? The clocks can never be removed and cleaning up on error paths is kind of pointless. The system would be hosed, right?

It depends - not all clocks are necessary for system to boot.

Quoted 2 lines

That doesn't matter. You have a single driver for all/most of the clocks, so the driver can't be removed.

-ECANOFWORMS

A couple of quick rebuttals, but I imagine we're going to disagree on the platform_driver thing as a matter of taste no matter what...

1) There should be multiple clk drivers in a properly modeled system. Some folks still incorrectly put all clocks in a system into a single driver because That's How We Used To Do It, and some systems (simpler ones) really only have a single clock generator IP block.

Excepting those two reasons above, we really should have separate drivers for clocks controlled by the PMIC, for the (one or more) clock generator blocks inside of the AP/SoC, and then even more for the drivers that map to IP blocks that have their own clock gens.

Good examples of the latter are display controllers that generate their own PLL and pixel clock. Or MMC controllers that have a runtime-programmable clock divider. Examples of these are merged into mainline.

2) Stephen and I wanted clock drivers to actually be represented in the driver model. There were these gigantic clock drivers that exclusively used CLK_OF_DECLARE and they just sort of floated out there in the ether... no representation in sysfs, no struct device to map onto a clock controller struct, etc.

I'd be happy to hear why you think that platform_driver is a bad fit for a device driver that generally manages memory-mapped system resources that are part of the system glue and not really tied to a specific bus. Sounds like a good fit to me.

If platform_driver doesn't handle the early boot thing well, that tells me that we have a problem to solve in platform_driver, not in the clk subsystem or drivers.

3) Lots of clock controllers should be loadable modules. E.g. i2c clock expanders, potentially external PMIC-related drivers, external audio codecs, etc.

Again, repeating my point #1 above, just because many platforms have a monolithic clock driver does not mean that this is the Right Way to do it.

Quoted 57 lines · Michael Turquette

Quoting Rob Herring (2018-05-14 06:20:57)

Quoted 49 lines · Bartosz Golaszewski

Quoted 48 lines

2018-05-11 22:13 GMT+02:00 Rob Herring <[email protected]>:

Quoted 15 lines · Bartosz Golaszewski

Quoted 5 lines

This series is a follow-up to the RFC[1] posted a couple days ago.

NOTE: this series applies on top of my recent patches[2] that move the previous implementation of early platform devices to arch/sh.

Quoted 8 lines

Problem:

Certain class of devices, such as timers, certain clock drivers and irq chip drivers need to be probed early in the boot sequence. The currently preferred approach is using one of the OF_DECLARE() macros. This however does not create a platform device which has many drawbacks - such as not being able to use devres routines, dev_ log functions or no way of deferring the init OF function if some other resources are missing.

Quoted 8 lines

I skimmed though this and it doesn't look horrible (how's that for positive feedback? ;) ). But before going into the details, I think first there needs to be agreement this is the right direction.

The question does remain though as to whether this class of devices should be platform drivers. They can't be modules. They can't be hotplugged. Can they be runtime-pm enabled? So the advantage is ...

The main (but not the only) advantage for drivers that can both be platform drivers and OF_DECLARE drivers is that we get a single entry point and can reuse code without resorting to checking if (!dev). It results in more consistent code base. Another big advantage is consolidation of device tree and machine code for SoC drivers used in different boards of which some are still using board files and others are defined in DT (see: DaVinci).

Quoted 14 lines

I assume that the clock maintainers had some reason to move clocks to be platform drivers. It's just not clear to me what that was.

Quoted 5 lines

For drivers that use both platform drivers and OF_DECLARE the situation is even more complicated as the code needs to take into account that there can possibly be no struct device present. For a specific use case that we're having problems with, please refer to the recent DaVinci common-clock conversion patches and the nasty workaround that this problem implies[3].

So devm_kzalloc will work with this solution? Why did we need devm_kzalloc in the first place? The clocks can never be removed and cleaning up on error paths is kind of pointless. The system would be hosed, right?

Quoted 4 lines

Quoted 1 line

It depends - not all clocks are necessary for system to boot.

That doesn't matter. You have a single driver for all/most of the clocks, so the driver can't be removed.

-ECANOFWORMS

Quoted 2 lines

A couple of quick rebuttals, but I imagine we're going to disagree on the platform_driver thing as a matter of taste no matter what...

It's really more should the clocksource, clockevents and primary interrupt controller be drivers. Let's get agreement on that first. If yes, then it probably does make sense that their dependencies are drivers too. If not, then making only the dependencies drivers doesn't seem right to me.

Quoted 9 lines

1) There should be multiple clk drivers in a properly modeled system. Some folks still incorrectly put all clocks in a system into a single driver because That's How We Used To Do It, and some systems (simpler ones) really only have a single clock generator IP block.

Excepting those two reasons above, we really should have separate drivers for clocks controlled by the PMIC, for the (one or more) clock generator blocks inside of the AP/SoC, and then even more for the drivers that map to IP blocks that have their own clock gens.

I agree those should be separate entities at least. But for a given h/w block, if you already have to use OF_DECLARE, why would you try to split that into OF_DECLARE and a driver? what advantage does putting non-boot essential clocks in a driver or transitioning to a driver get you?

And I've seen PMIC clocks could be inputs to the SoC's clock controller(s), so the dependencies get more complicated. Then does the PMIC driver and its dependencies need to be early drivers too?

Quoted 4 lines

Good examples of the latter are display controllers that generate their own PLL and pixel clock. Or MMC controllers that have a runtime-programmable clock divider. Examples of these are merged into mainline.

But those are drivers of types other than a clock controller that happen to register some clocks as well. I wasn't saying these cases can't or shouldn't be part of the driver model. Look at irqchips. We have some that use the driver model (e.g. every GPIO driver) and some that don't because there's no need (e.g. GIC).

Quoted 11 lines

2) Stephen and I wanted clock drivers to actually be represented in the driver model. There were these gigantic clock drivers that exclusively used CLK_OF_DECLARE and they just sort of floated out there in the ether... no representation in sysfs, no struct device to map onto a clock controller struct, etc.

I'd be happy to hear why you think that platform_driver is a bad fit for a device driver that generally manages memory-mapped system resources that are part of the system glue and not really tied to a specific bus. Sounds like a good fit to me.

Quoted 3 lines

If platform_driver doesn't handle the early boot thing well, that tells me that we have a problem to solve in platform_driver, not in the clk subsystem or drivers.

Doing things earlier is not the only way to solve the problems. Perhaps we need to figure out how to start things later. Maybe it's not feasible here, I don't know.

Yes, they should all be drivers.

Assuming clocksources, clockevents, or primary interrupt controllers are special, and thus forcing everyone to use OF_DECLARE for the whole subsystem, doesn't fly everywhere.

Clockevents and interrupt controllers can have a module clock. All three can be part of a PM Domain, which requires a struct device to be handled properly.

This series is a follow-up to the RFC[1] posted a couple days ago.

NOTE: this series applies on top of my recent patches[2] that move the previous implementation of early platform devices to arch/sh.

Problem:

Certain class of devices, such as timers, certain clock drivers and irq chip drivers need to be probed early in the boot sequence. The currently preferred approach is using one of the OF_DECLARE() macros. This however does not create a platform device which has many drawbacks - such as not being able to use devres routines, dev_ log functions or no way of deferring the init OF function if some other resources are missing.

For drivers that use both platform drivers and OF_DECLARE the situation is even more complicated as the code needs to take into account that there can possibly be no struct device present. For a specific use case that we're having problems with, please refer to the recent DaVinci common-clock conversion patches and the nasty workaround that this problem implies[3].

We also used to have an early platform drivers implementation but they were not integrated with the linux device model at all - they merely used the same data structures. The users could not use devres, defer probe and the early devices never became actual platform devices later on.

Proposed solution:

This series aims at solving this problem by (re-)introducing the concept of early platform drivers and devices - this time however in a way that seamlessly integrates with the existing platform drivers and also offers device-tree support.

The idea is to provide a way for users to probe devices early, while already being able to use devres, devices resources and properties and also deferred probing.

New structures are introduced: the early platform driver contains the early_probe callback which has the same signature as regular platform_device probe. This callback is called early on. The user can have both the early and regular probe speficied or only one of them and they both receive the same platform device object as argument. Any device data allocated early will be carried over to the normal probe.

The architecture code is responsible for calling early_platform_start() in which the early drivers will be registered and devices populated from DT.

Once the device and kobject mechanisms are ready, all early drivers and devices will be converted into real platform drivers and devices. Also: if any of the early platform registration functions will be called once early initialization is done, these functions will work like regular platform_device/driver ones.

Patches 1-9/12 introduce changes to existing code that are necessary before adding support for early drivers. Patch 10/12 contains the new framwork in an isolated file which can be compiled only if needed by the architecture.

Patch 11/12 contains a dummy early platform driver that serves as a code example and can be used for simple testing.

The last patch finally makes the of/platform code aware of early platform drivers.

If accepted, this new mechanism could potentially lead to consolidation of the code currently used by users of OF_DECLARE, since they could be converted to actual platform drivers.

[1] https://lkml.org/lkml/2018/4/26/657
[2] https://lkml.org/lkml/2018/4/30/547
[3] https://lkml.org/lkml/2018/4/26/1094

Bartosz Golaszewski (12):
platform/early: add a new field to struct device platform/early: don't WARN() on non-empty devres list for early
devices
platform/early: export platform_match() locally
platform: provide a separate function for initializing platform
devices
platform: export platform_device_release() locally
of: add a new flag for OF device nodes
of/platform: provide a separate routine for setting up device
resources
of/platform: provide a separate routine for device initialization platform/early: add an init section for early driver data platform/early: implement support for early platform drivers
misc: implement a dummy early platform driver of/platform: make the OF code aware of early platform drivers

 drivers/base/Kconfig              |   3 +
 drivers/base/Makefile             |   1 +
 drivers/base/base.h               |   4 +
 drivers/base/dd.c                 |   2 +-
 drivers/base/early.c              | 332 ++++++++++++++++++++++++++++++
 drivers/base/platform.c           |  28 ++-
 drivers/misc/Kconfig              |   8 +
 drivers/misc/Makefile             |   2 +
 drivers/misc/dummy-early.c        |  82 ++++++++
 drivers/of/platform.c             |  85 +++++---
 include/asm-generic/vmlinux.lds.h |  11 +
 include/linux/device.h            |   1 +
 include/linux/early_platform.h    |  75 +++++++
 include/linux/of.h                |   1 +
 include/linux/of_platform.h       |   2 +
 include/linux/platform_device.h   |   2 +
 16 files changed, 604 insertions(+), 35 deletions(-)
 create mode 100644 drivers/base/early.c
 create mode 100644 drivers/misc/dummy-early.c
 create mode 100644 include/linux/early_platform.h

--
2.17.0