arm64: change ARCH_SPRD Kconfig to tristate
Subsystems (2)
PATCH 1/2 arm64: change ARCH_SPRD Kconfig to tristate
From: Chunyan Zhang <[email protected]>
The default value of Kconfig for almost all sprd drivers are the same with ARCH_SPRD, making these drivers built as modules as default would be easier if we can set ARCH_SPRD as 'm', so this patch change ARCH_SPRD to tristate.
--- arch/arm64/Kconfig.platforms | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/Kconfig.platforms +1 -1
@@ -249,7 +249,7 @@ config ARCH_TEGRA This enables support for the NVIDIA Tegra SoC family. config ARCH_SPRD - bool "Spreadtrum SoC platform" + tristate "Spreadtrum SoC platform" help Support for Spreadtrum ARM based SoCs
Signature · 1 line
2.20.1
PATCH 2/2 tty: serial: make SERIAL_SPRD not depend on ARCH_SPRD
From: Chunyan Zhang <[email protected]>
Remove the dependency with ARCH_SPRD from sprd serial/console Kconfig-s, since we want them can be built-in when ARCH_SPRD is set as 'm'.
--- drivers/tty/serial/Kconfig | 1 - 1 file changed, 1 deletion(-)
drivers/tty/serial/Kconfig +0 -1
@@ -1452,7 +1452,6 @@ config SERIAL_MEN_Z135 config SERIAL_SPRD tristate "Support for Spreadtrum serial" - depends on ARCH_SPRD select SERIAL_CORE help This enables the driver for the Spreadtrum's serial.
Signature · 1 line
2.20.1
Re: [PATCH 2/2] tty: serial: make SERIAL_SPRD not depend on ARCH_SPRD
Hi Chunyan,
Quoted 4 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
Remove the dependency with
ARCH_SPRDfrom sprd serial/console Kconfig-s, since we want them can be built-in whenARCH_SPRDis set as 'm'.
Why would you want a serial driver for a specific platform to be builtin, while all other platform support is modular?
Quoted 10 lines
Signed-off-by Chunyan Zhang <[email protected]>--- drivers/tty/serial/Kconfig | 1 - 1 file changed, 1 deletion(-)drivers/tty/serial/Kconfig +0
-0
drivers/tty/serial/Kconfig config SERIAL_MEN_Z135 · line 1457
Quoted 7 lines
drivers/tty/serial/Kconfig +0
-1@@ -1452,7 +1452,6 @@ config SERIAL_MEN_Z135 config SERIAL_SPRD tristate "Support for Spreadtrum serial" - depends on ARCH_SPRD select SERIAL_CORE help This enables the driver for the Spreadtrum's serial.
Gr{oetje,eeting}s,
Geert
Signature · 5 lines
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
Re: [PATCH 2/2] tty: serial: make SERIAL_SPRD not depend on ARCH_SPRD
Hi Geert,
Quoted 11 lines · Geert Uytterhoeven
Hi Chunyan,
Quoted 4 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
Remove the dependency with
ARCH_SPRDfrom sprd serial/console Kconfig-s, since we want them can be built-in whenARCH_SPRDis set as 'm'.Why would you want a serial driver for a specific platform to be builtin, while all other platform support is modular?
Oh, that's not this patch means.
We just want serial driver can be builtin for any platform, so it should not depend on a config which can be set as 'm' (i.e. ARCH_SPRD) , otherwise if the config was set as 'm', the serial driver can't be selected as 'y' then.
That's what I mean.
Cheers,
Chunyan
Quoted 29 lines
Quoted 11 lines
Signed-off-by Chunyan Zhang <[email protected]>--- drivers/tty/serial/Kconfig | 1 - 1 file changed, 1 deletion(-)drivers/tty/serial/Kconfig +0
-0drivers/tty/serial/Kconfig config SERIAL_MEN_Z135 · line 1457
Quoted 7 lines
drivers/tty/serial/Kconfig +0
-1@@ -1452,7 +1452,6 @@ config SERIAL_MEN_Z135 config SERIAL_SPRD tristate "Support for Spreadtrum serial" - depends on ARCH_SPRD select SERIAL_CORE help This enables the driver for the Spreadtrum's serial.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
Re: [PATCH 2/2] tty: serial: make SERIAL_SPRD not depend on ARCH_SPRD
Hi Chunyan,
Quoted 7 lines · Chunyan Zhang
Quoted 6 lines · Geert Uytterhoeven
Quoted 4 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
Remove the dependency with
ARCH_SPRDfrom sprd serial/console Kconfig-s, since we want them can be built-in whenARCH_SPRDis set as 'm'.
Quoted 6 lines
Quoted 2 lines
Why would you want a serial driver for a specific platform to be builtin, while all other platform support is modular?
Oh, that's not this patch means.
We just want serial driver can be builtin for any platform, so it
What would be the benefit of the user to be able to have the SPRD serial driver built-in on any platform? AFAIU, it supports only Spreadtrum platforms.
Quoted 3 lines
should not depend on a config which can be set as 'm' (i.e.
ARCH_SPRD) , otherwise if the config was set as 'm', the serial driver can't be selected as 'y' then.
I ask about that as a reply to PATCH 1/2.
Quoted 6 lines
That's what I mean.
Quoted 4 lines
Quoted 4 lines
drivers/tty/serial/Kconfig +0
-0@@ -1452,7 +1452,6 @@ config SERIAL_MEN_Z135
Quoted 6 lines
Quoted 6 lines
drivers/tty/serial/Kconfig config SERIAL_MEN_Z135 · line 1457
Quoted 6 lines
drivers/tty/serial/Kconfig +0
-1config SERIAL_SPRD tristate "Support for Spreadtrum serial" - depends on ARCH_SPRD select SERIAL_CORE help This enables the driver for the Spreadtrum's serial.
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
Re: [PATCH 2/2] tty: serial: make SERIAL_SPRD not depend on ARCH_SPRD
Hi Geert,
Quoted 14 lines · Geert Uytterhoeven
Hi Chunyan,
Quoted 10 lines · Chunyan Zhang
Quoted 8 lines · Geert Uytterhoeven
Quoted 4 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
Remove the dependency with
ARCH_SPRDfrom sprd serial/console Kconfig-s, since we want them can be built-in whenARCH_SPRDis set as 'm'.Why would you want a serial driver for a specific platform to be builtin, while all other platform support is modular?
Quoted 7 lines
Quoted 3 lines
Oh, that's not this patch means.
We just want serial driver can be builtin for any platform, so it
What would be the benefit of the user to be able to have the SPRD serial driver built-in on any platform? AFAIU, it supports only Spreadtrum platforms.
Right, it does support Spreadtrum platforms only indeed. Like I said on the replay to patch 1/2, simply because I want serial driver can be builtin all the time, no matter ARCH_SPRD is m or y.
Regards,
Chunyan
Quoted 30 lines
Quoted 4 lines
should not depend on a config which can be set as 'm' (i.e.
ARCH_SPRD) , otherwise if the config was set as 'm', the serial driver can't be selected as 'y' then.I ask about that as a reply to PATCH 1/2.
Quoted 6 lines
That's what I mean.
Quoted 4 lines
Quoted 4 lines
drivers/tty/serial/Kconfig +0
-0@@ -1452,7 +1452,6 @@ config SERIAL_MEN_Z135Quoted 6 lines
Quoted 6 lines
drivers/tty/serial/Kconfig config SERIAL_MEN_Z135 · line 1457
Quoted 6 lines
drivers/tty/serial/Kconfig +0
-1config SERIAL_SPRD tristate "Support for Spreadtrum serial" - depends on ARCH_SPRD select SERIAL_CORE help This enables the driver for the Spreadtrum's serial.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
Re: [PATCH 2/2] tty: serial: make SERIAL_SPRD not depend on ARCH_SPRD
Hi Chunyan,
Quoted 16 lines · Chunyan Zhang
Quoted 15 lines · Geert Uytterhoeven
Quoted 13 lines · Chunyan Zhang
Quoted 8 lines · Geert Uytterhoeven
Quoted 4 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
Remove the dependency with
ARCH_SPRDfrom sprd serial/console Kconfig-s, since we want them can be built-in whenARCH_SPRDis set as 'm'.Why would you want a serial driver for a specific platform to be builtin, while all other platform support is modular?
Oh, that's not this patch means.
We just want serial driver can be builtin for any platform, so it
Quoted 7 lines
Quoted 3 lines
What would be the benefit of the user to be able to have the SPRD serial driver built-in on any platform? AFAIU, it supports only Spreadtrum platforms.
Right, it does support Spreadtrum platforms only indeed. Like I said on the replay to patch 1/2, simply because I want serial driver can be builtin all the time, no matter
ARCH_SPRDis m or y.
OK.
So shouldn't the dependency become
depends on ARCH_SPRD || ARCH_SPRD=m || COMPILE_TEST
instead, to avoid asking the question when you're not building a kernel plus modules for Spreadtrum platforms?
Quoted 12 lines
Quoted 12 lines
Quoted 3 lines
should not depend on a config which can be set as 'm' (i.e.
ARCH_SPRD) , otherwise if the config was set as 'm', the serial driver can't be selected as 'y' then.I ask about that as a reply to PATCH 1/2.
Quoted 6 lines
That's what I mean.
Quoted 4 lines
Quoted 4 lines
drivers/tty/serial/Kconfig +0
-0@@ -1452,7 +1452,6 @@ config SERIAL_MEN_Z135
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
drivers/tty/serial/Kconfig config SERIAL_MEN_Z135 · line 1457
Quoted 6 lines
drivers/tty/serial/Kconfig +0
-1config SERIAL_SPRD tristate "Support for Spreadtrum serial" - depends on ARCH_SPRD select SERIAL_CORE help This enables the driver for the Spreadtrum's serial.
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
Re: [PATCH 2/2] tty: serial: make SERIAL_SPRD not depend on ARCH_SPRD
Quoted 28 lines · Geert Uytterhoeven
Hi Chunyan,
Quoted 16 lines · Chunyan Zhang
Quoted 15 lines · Geert Uytterhoeven
Quoted 13 lines · Chunyan Zhang
Quoted 8 lines · Geert Uytterhoeven
Quoted 4 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
Remove the dependency with
ARCH_SPRDfrom sprd serial/console Kconfig-s, since we want them can be built-in whenARCH_SPRDis set as 'm'.Why would you want a serial driver for a specific platform to be builtin, while all other platform support is modular?
Oh, that's not this patch means.
We just want serial driver can be builtin for any platform, so it
Quoted 7 lines
Quoted 3 lines
What would be the benefit of the user to be able to have the SPRD serial driver built-in on any platform? AFAIU, it supports only Spreadtrum platforms.
Right, it does support Spreadtrum platforms only indeed. Like I said on the replay to patch 1/2, simply because I want serial driver can be builtin all the time, no matter
ARCH_SPRDis m or y.
Quoted 5 lines
OK.
So shouldn't the dependency become
depends on ARCH_SPRD || ARCH_SPRD=m || COMPILE_TEST
Oh, right, this's better than just removing dependency to ARCH_SPRD, but considering that this patch has been merged into Greg's tree, I will post another patch to add this.
Thanks for your comments!
Chunyan
Quoted 34 lines
instead, to avoid asking the question when you're not building a kernel plus modules for Spreadtrum platforms?
Quoted 12 lines
Quoted 12 lines
Quoted 3 lines
should not depend on a config which can be set as 'm' (i.e.
ARCH_SPRD) , otherwise if the config was set as 'm', the serial driver can't be selected as 'y' then.I ask about that as a reply to PATCH 1/2.
Quoted 6 lines
That's what I mean.
Quoted 4 lines
Quoted 4 lines
drivers/tty/serial/Kconfig +0
-0@@ -1452,7 +1452,6 @@ config SERIAL_MEN_Z135Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
drivers/tty/serial/Kconfig config SERIAL_MEN_Z135 · line 1457
Quoted 6 lines
drivers/tty/serial/Kconfig +0
-1config SERIAL_SPRD tristate "Support for Spreadtrum serial" - depends on ARCH_SPRD select SERIAL_CORE help This enables the driver for the Spreadtrum's serial.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
Re: [PATCH 2/2] tty: serial: make SERIAL_SPRD not depend on ARCH_SPRD
In reply to Geert UytterhoevenHi Geert,
Quoted 14 lines · Geert Uytterhoeven
Hi Chunyan,
Quoted 10 lines · Chunyan Zhang
Quoted 8 lines · Geert Uytterhoeven
Quoted 4 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
Remove the dependency with
ARCH_SPRDfrom sprd serial/console Kconfig-s, since we want them can be built-in whenARCH_SPRDis set as 'm'.Why would you want a serial driver for a specific platform to be builtin, while all other platform support is modular?
Quoted 8 lines
Quoted 3 lines
Oh, that's not this patch means.
We just want serial driver can be builtin for any platform, so it
What would be the benefit of the user to be able to have the SPRD serial driver built-in on any platform? AFAIU, it supports only Spreadtrum platforms.
Assume there is a all-in-one common kernel binary image say arm64 for all SoC platform and all
device drivers are built as modules from different vendors for each specific devices.
But for serial driver it is too late to be initialized as a module if users want to check early console log.
At this circumstance vendors prefer to set their serial driver to be built into that all-in-one image as easy
as to only set one config instead of enabling whole platform then disabling many out of serial.
-Orson
Quoted 29 lines
Quoted 3 lines
should not depend on a config which can be set as 'm' (i.e.
ARCH_SPRD) , otherwise if the config was set as 'm', the serial driver can't be selected as 'y' then.I ask about that as a reply to PATCH 1/2.
Quoted 6 lines
That's what I mean.
Quoted 4 lines
Quoted 4 lines
drivers/tty/serial/Kconfig +0
-0@@ -1452,7 +1452,6 @@ config SERIAL_MEN_Z135Quoted 6 lines
Quoted 6 lines
drivers/tty/serial/Kconfig config SERIAL_MEN_Z135 · line 1457
Quoted 6 lines
drivers/tty/serial/Kconfig +0
-1config SERIAL_SPRD tristate "Support for Spreadtrum serial" - depends on ARCH_SPRD select SERIAL_CORE help This enables the driver for the Spreadtrum's serial.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
Hi Chunyan.
Quoted 7 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
The default value of
Kconfigfor almost all sprd drivers are the same withARCH_SPRD, making these drivers built as modules as default would be easier if we can setARCH_SPRDas 'm', so this patch changeARCH_SPRDto tristate.Signed-off-by Chunyan Zhang <[email protected]>
Can you actually boot a kernel on a Spreadtrum platform when all platform and driver support is modular?
arch/arm64/Kconfig.platforms config ARCH_TEGRA · line 254
Quoted 9 lines
+++ b/arch/arm64/Kconfig.platforms
arch/arm64/Kconfig.platforms +1
-1@@ -249,7 +249,7 @@ config ARCH_TEGRA This enables support for the NVIDIA Tegra SoC family. config ARCH_SPRD - bool "Spreadtrum SoC platform" + tristate "Spreadtrum SoC platform" help Support for Spreadtrum ARM based SoCs
Gr{oetje,eeting}s,
Geert
Signature · 5 lines
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
Hi Geert,
Quoted 6 lines · Geert Uytterhoeven
Hi Chunyan.
Quoted 2 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
Quoted 8 lines
Quoted 5 lines
The default value of
Kconfigfor almost all sprd drivers are the same withARCH_SPRD, making these drivers built as modules as default would be easier if we can setARCH_SPRDas 'm', so this patch changeARCH_SPRDto tristate.Signed-off-by Chunyan Zhang <[email protected]>Can you actually boot a kernel on a Spreadtrum platform when all platform and driver support is modular?
Yes, even if all drivers are modular.
But I hope serial can be builtin, then I can have a console to see kernel output before loading modules.
Also, this's what Google GKI [1] asked :)
Regards,
Chunyan
Quoted 13 lines
arch/arm64/Kconfig.platforms config ARCH_TEGRA · line 254
Quoted 10 lines
+++ b/arch/arm64/Kconfig.platforms
arch/arm64/Kconfig.platforms +1
-1@@ -249,7 +249,7 @@ config ARCH_TEGRA This enables support for the NVIDIA Tegra SoC family. config ARCH_SPRD - bool "Spreadtrum SoC platform" + tristate "Spreadtrum SoC platform" help Support for Spreadtrum ARM based SoCsGr{oetje,eeting}s,
Quoted 8 lines
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
_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Hi Chunyan,
Quoted 8 lines · Chunyan Zhang
Quoted 7 lines · Geert Uytterhoeven
Quoted 6 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
The default value of
Kconfigfor almost all sprd drivers are the same withARCH_SPRD, making these drivers built as modules as default would be easier if we can setARCH_SPRDas 'm', so this patch changeARCH_SPRDto tristate.
Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Signed-off-by Chunyan Zhang <[email protected]>Can you actually boot a kernel on a Spreadtrum platform when all platform and driver support is modular?
Yes, even if all drivers are modular.
Cool. No hard dependencies on e.g. regulators that are turned off when unused?
Quoted 2 lines
But I hope serial can be builtin, then I can have a console to see kernel output before loading modules.
No dependency on the clock driver?
Oh, I see you have a hack in the serial driver, to assume default values when the serial port's parent clock is not found. That may limit use of the other serial ports, depending on the actual serial hardware.
And on Sharkl64, the serial port's clock is a fixed-clock anyway, so you don't even need the hack.
But in general you cannot rely on that, especially if your SoC has clock and/or power domains.
BTW, what about the watchdog driver? That one does need a clock, and loading it too late will reboot your system.
Quoted 6 lines
Also, this's what Google GKI [1] asked :)
Regards,
Chunyan
Let's see how having everything modular works out on an SoC where all hardware is part of a clock and power domain.
Thanks!
Gr{oetje,eeting}s,
Geert
Signature · 5 lines
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
Hi Geert,
Quoted 40 lines · Geert Uytterhoeven
Hi Chunyan,
Quoted 8 lines · Chunyan Zhang
Quoted 7 lines · Geert Uytterhoeven
Quoted 6 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
The default value of
Kconfigfor almost all sprd drivers are the same withARCH_SPRD, making these drivers built as modules as default would be easier if we can setARCH_SPRDas 'm', so this patch changeARCH_SPRDto tristate.Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Signed-off-by Chunyan Zhang <[email protected]>Can you actually boot a kernel on a Spreadtrum platform when all platform and driver support is modular?
Yes, even if all drivers are modular.
Cool. No hard dependencies on e.g. regulators that are turned off when unused?
Quoted 2 lines
But I hope serial can be builtin, then I can have a console to see kernel output before loading modules.
No dependency on the clock driver?
Oh, I see you have a hack in the serial driver, to assume default values when the serial port's parent clock is not found. That may limit use of the other serial ports, depending on the actual serial hardware.
And on Sharkl64, the serial port's clock is a fixed-clock anyway, so you don't even need the hack.But in general you cannot rely on that, especially if your SoC has clock and/or power domains.
BTW, what about the watchdog driver? That one does need a clock, and loading it too late will reboot your system.
Quoted 2 lines
Also, this's what Google GKI [1] asked :)
Quoted 7 lines
Quoted 4 lines
Regards,
ChunyanLet's see how having everything modular works out on an SoC where all hardware is part of a clock and power domain.
I'm curious, are there any problems that we should be aware of? We know about the regulator sync state and consumer-supplier dependencies. [1]
(Adding Saravana inline)
Thanks,
- ssp
Quoted 48 lines · Sandeep Patil
Hi Geert,
Quoted 45 lines · Geert Uytterhoeven
Hi Chunyan,
Quoted 8 lines · Chunyan Zhang
Quoted 7 lines · Geert Uytterhoeven
Quoted 6 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
The default value of
Kconfigfor almost all sprd drivers are the same withARCH_SPRD, making these drivers built as modules as default would be easier if we can setARCH_SPRDas 'm', so this patch changeARCH_SPRDto tristate.Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Signed-off-by Chunyan Zhang <[email protected]>Can you actually boot a kernel on a Spreadtrum platform when all platform and driver support is modular?
Yes, even if all drivers are modular.
Cool. No hard dependencies on e.g. regulators that are turned off when unused?
Quoted 2 lines
But I hope serial can be builtin, then I can have a console to see kernel output before loading modules.
No dependency on the clock driver?
Oh, I see you have a hack in the serial driver, to assume default values when the serial port's parent clock is not found. That may limit use of the other serial ports, depending on the actual serial hardware.
And on Sharkl64, the serial port's clock is a fixed-clock anyway, so you don't even need the hack.But in general you cannot rely on that, especially if your SoC has clock and/or power domains.
BTW, what about the watchdog driver? That one does need a clock, and loading it too late will reboot your system.
Quoted 6 lines
Also, this's what Google GKI [1] asked :)
Regards,
Chunyan
Quoted 8 lines
Quoted 2 lines
Let's see how having everything modular works out on an SoC where all hardware is part of a clock and power domain.
I'm curious, are there any problems that we should be aware of? We know about the regulator sync state and consumer-supplier dependencies. [1]
(Adding Saravana inline)
(oops, forgot to paste the link to presentation)
1. https://linuxplumbersconf.org/event/4/contributions/402/attachments/320/544/Solving_issues_associated_with_modules_and_supplier-consumer_dependencies.pdf
Hi Sandeep,
Quoted 32 lines · Sandeep Patil
Quoted 31 lines · Sandeep Patil
Quoted 30 lines · Geert Uytterhoeven
Quoted 8 lines · Chunyan Zhang
Quoted 7 lines · Geert Uytterhoeven
Quoted 6 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
The default value of
Kconfigfor almost all sprd drivers are the same withARCH_SPRD, making these drivers built as modules as default would be easier if we can setARCH_SPRDas 'm', so this patch changeARCH_SPRDto tristate.Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Signed-off-by Chunyan Zhang <[email protected]>Can you actually boot a kernel on a Spreadtrum platform when all platform and driver support is modular?
Yes, even if all drivers are modular.
Cool. No hard dependencies on e.g. regulators that are turned off when unused?
Quoted 2 lines
But I hope serial can be builtin, then I can have a console to see kernel output before loading modules.
No dependency on the clock driver?
Oh, I see you have a hack in the serial driver, to assume default values when the serial port's parent clock is not found. That may limit use of the other serial ports, depending on the actual serial hardware.
And on Sharkl64, the serial port's clock is a fixed-clock anyway, so you don't even need the hack.
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
But in general you cannot rely on that, especially if your SoC has clock and/or power domains.
So I gave it a try on my platform (R-Car), with just the serial driver builtin. I can see the kernel booting using earlycon. However, the serial driver is never probed, as the pin control driver is missing. And after that, it would need the clock domain (power domain and clock drivers), and more serial clock inputs (clock driver, again). Fortunately DMA is optional (dmac driver, and optional iommu driver).
I guess core platform-specific drivers can be loaded from the initramfs, after which the serial driver would be probed, finally. If the initramfs includes a storage driver, anything else can be loaded from e.g. eMMC.
Quoted 9 lines
Quoted 9 lines
Quoted 6 lines
Quoted 3 lines
Also, this's what Google GKI [1] asked :)
Let's see how having everything modular works out on an SoC where all hardware is part of a clock and power domain.
I'm curious, are there any problems that we should be aware of? We know about the regulator sync state and consumer-supplier dependencies. [1]
Care must be taken, as probe deferral behaves differently after late_initcall time (see driver_deferred_probe_check_state_continue), which may cause issues with "optional" components like DMA controllers and IOMMUs. Unused clocks and power domains are not turned off, unlike for the builtin case.
On some SoCs, even the interrupt controller (GIC) may be part of a clock and/or power domain, while the GIC driver is not aware of that. For the clock, that is handled by CLK_IS_CRITICAL in the clock driver. For power domains, that is not handled explicitly on Renesas SoCs, as the GIC is always in the same power domain as other devices that are Runtime PM-aware. But if those devices would have modular drivers, that power domain could still be turned off.
So on a modern ARM system, I assume a generic kernel with no vendor-specific drivers builtin can indeed work.
On older systems, without an ARM architectured timer, you will need at least the platform-specific timer driver (and the clock driver) builtin. Assumed the timer's power domain has been powered up by the firmware.
Quoted 2 lines
(oops, forgot to paste the link to presentation)
1. https://linuxplumbersconf.org/event/4/contributions/402/attachments/320/544/Solving_issues_associated_with_modules_and_supplier-consumer_dependencies.pdf
Thanks, hadn't seen that presentation.
It matches my understanding.
Gr{oetje,eeting}s,
Geert
Signature · 5 lines
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
Hi Geert,
Quoted 22 lines · Geert Uytterhoeven
Hi Chunyan,
Quoted 8 lines · Chunyan Zhang
Quoted 7 lines · Geert Uytterhoeven
Quoted 6 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
The default value of
Kconfigfor almost all sprd drivers are the same withARCH_SPRD, making these drivers built as modules as default would be easier if we can setARCH_SPRDas 'm', so this patch changeARCH_SPRDto tristate.Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Signed-off-by Chunyan Zhang <[email protected]>Can you actually boot a kernel on a Spreadtrum platform when all platform and driver support is modular?
Yes, even if all drivers are modular.
Cool. No hard dependencies on e.g. regulators that are turned off when unused?
Quoted 8 lines
Quoted 2 lines
But I hope serial can be builtin, then I can have a console to see kernel output before loading modules.
No dependency on the clock driver?
Oh, I see you have a hack in the serial driver, to assume default values when the serial port's parent clock is not found. That may limit use of the other serial ports, depending on the actual serial hardware.
There is an function named "sprd_uart_is_console()" in the driver code. So the hack could be only applied when the
port is identified as console. And other ports might return PROBE_DEFER until the clock is ready.
Could it work out of the limitation?
-Orson
Quoted 31 lines
And on Sharkl64, the serial port's clock is a fixed-clock anyway, so you don't even need the hack.
But in general you cannot rely on that, especially if your SoC has clock and/or power domains.
BTW, what about the watchdog driver? That one does need a clock, and loading it too late will reboot your system.
Quoted 6 lines
Also, this's what Google GKI [1] asked :)
Regards,
ChunyanLet's see how having everything modular works out on an SoC where all hardware is part of a clock and power domain.
Thanks!
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
Hi Orson,
Quoted 29 lines · Orson Zhai
Quoted 19 lines · Geert Uytterhoeven
Quoted 8 lines · Chunyan Zhang
Quoted 7 lines · Geert Uytterhoeven
Quoted 6 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
The default value of
Kconfigfor almost all sprd drivers are the same withARCH_SPRD, making these drivers built as modules as default would be easier if we can setARCH_SPRDas 'm', so this patch changeARCH_SPRDto tristate.Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Signed-off-by Chunyan Zhang <[email protected]>Can you actually boot a kernel on a Spreadtrum platform when all platform and driver support is modular?
Yes, even if all drivers are modular.
Cool. No hard dependencies on e.g. regulators that are turned off when unused?
Quoted 8 lines
Quoted 2 lines
But I hope serial can be builtin, then I can have a console to see kernel output before loading modules.
No dependency on the clock driver?
Oh, I see you have a hack in the serial driver, to assume default values when the serial port's parent clock is not found. That may limit use of the other serial ports, depending on the actual serial hardware.
Quoted 6 lines
There is an function named "
sprd_uart_is_console()" in the driver code. So the hack could be only applied when the
port is identified as console. And other ports might returnPROBE_DEFERuntil the clock is ready.Could it work out of the limitation?
Yes, that could work. You also have only a single SPRD_DEFAULT_SOURCE_CLK, which makes it simple to handle.
For other SoCs, there may be a variation of possible values, depending on SoC and/or board.
Gr{oetje,eeting}s,
Geert
Signature · 5 lines
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
Quoted 38 lines · Geert Uytterhoeven
Hi Orson,
Quoted 34 lines · Orson Zhai
Quoted 19 lines · Geert Uytterhoeven
Quoted 8 lines · Chunyan Zhang
Quoted 7 lines · Geert Uytterhoeven
Quoted 6 lines · Chunyan Zhang
From: Chunyan Zhang <[email protected]>
The default value of
Kconfigfor almost all sprd drivers are the same withARCH_SPRD, making these drivers built as modules as default would be easier if we can setARCH_SPRDas 'm', so this patch changeARCH_SPRDto tristate.Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Signed-off-by Chunyan Zhang <[email protected]>Can you actually boot a kernel on a Spreadtrum platform when all platform and driver support is modular?
Yes, even if all drivers are modular.
Cool. No hard dependencies on e.g. regulators that are turned off when unused?
Quoted 8 lines
Quoted 2 lines
But I hope serial can be builtin, then I can have a console to see kernel output before loading modules.
No dependency on the clock driver?
Oh, I see you have a hack in the serial driver, to assume default values when the serial port's parent clock is not found. That may limit use of the other serial ports, depending on the actual serial hardware.There is an function named "
sprd_uart_is_console()" in the driver code. So the hack could be only applied when the
port is identified as console. And other ports might returnPROBE_DEFERuntil the clock is ready.
Quoted 6 lines
Quoted 1 line
Could it work out of the limitation?
Yes, that could work. You also have only a single
SPRD_DEFAULT_SOURCE_CLK, which makes it simple to handle.
For other SoCs, there may be a variation of possible values, depending on SoC and/or board.
Ok, thanks.
-Orson
Quoted 11 lines
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