Add support for RPMI to U-Boot
Subsystems (2)
cover Add support for RPMI to U-Boot
Hello,
This series adds support for RISC-V Platform Management Interface (RPMI) to U-Boot. RPMI is an OS-agnostic protocol for communication between an Application Processor (AP) and a Platform Microcontroller (PuC) [1]. The goals and purpose of RPMI are similar to ARM's SCMI.
This patchset first adds an abstraction layer for the transport (UCLASS_RPMI), then adds two transport layer (SBI MPXY and shared memory) and finally adds two client, or service group in RPMI jargon: Clock and device power service groups.
This series also include a sandbox mode for which the RPMI transport doesn't attach to a firmware but contains a simulation of a firmware. This reuses code from librpmi [2] to create the message responses.
Apart from the spec [1], this series reuses code from the Linux SBI MPXY driver [3], OpenSBI shared memory transport [4] and Linux RPMI clock driver [5].
This was tested on a PIC64-HPSC SoC which features an application cluster and a system controller with isolated address spaces, hence the need for RPMI. I've tested the interoperability between RPMI clients and transports with an S-Mode U-boot (using the MPXY transport) and M-Mode U-Boot (using shared memory transport) and the following device tree fragment:
```
firmware {
rpmi: mailbox@7000000000 {
compatible = "riscv,rpmi-shmem-mbox";
reg = <0x70 0x00000000 0x0 0x3000>,
<0x70 0x00003000 0x0 0x3000>,
<0x70 0x00006000 0x0 0x1000>,
<0x70 0x00007000 0x0 0x1000>;
reg-names = "a2p-req", "p2a-ack", "p2a-req", "a2p-ack";
riscv,slot-size = <64>;
#mbox-cells = <1>;
};
ifndef CONFIG_RISCV_MMODE
clock-service {
compatible = "riscv,rpmi-mpxy-clock";
mboxes = <&rpmi 0x8>;
riscv,sbi-mpxy-channel-id = <0x1000>;
};
device-power-service {
compatible = "riscv,rpmi-mpxy-device-power";
mboxes = <&rpmi 0x9>;
riscv,sbi-mpxy-channel-id = <0x1001>;
};
mpxy_mbox: mailbox {
compatible = "riscv,sbi-mpxy-mbox";
#mbox-cells = <2>;
};
endif
sysc_clk: clock-controller {
compatible = "riscv,rpmi-clock";
ifdef CONFIG_RISCV_MMODE
mboxes = <&rpmi 0x8>;
else
mboxes = <&mpxy_mbox 0x1000 0x0>;
endif
#clock-cells = <1>;
};
sysc_dpwr: power-domain-controller {
compatible = "riscv,rpmi-device-power";
ifdef CONFIG_RISCV_MMODE
mboxes = <&rpmi 0x9>;
else
mboxes = <&mpxy_mbox 0x1001 0x0>;
endif
#power-domain-cells = <1>;
};
};
```[1]: https://github.com/riscv-non-isa/riscv-rpmi
[2]: https://github.com/riscv-software-src/librpmi
[3]: https://elixir.bootlin.com/linux/v7.0.10/source/drivers/mailbox/riscv-sbi-mpxy-mbox.c
[4]: https://elixir.bootlin.com/opensbi/v1.8.1/source/lib/utils/mailbox/fdt_mailbox_rpmi_shmem.c
[5]: https://elixir.bootlin.com/linux/v7.0.10/source/drivers/clk/clk-rpmi.c
Charles Perry (7):
firmware: add support for RPMI
firmware: rpmi: add support for the SBI MPXY transport
firmware: rpmi: add support for shared memory transport
drivers: clk: add support for RPMI clocks
drivers: power: add support for RPMI power domains
firmware: rpmi: add a test and a sandbox driver
firmware: rpmi: add a test and sandbox for device power
MAINTAINERS | 9 + arch/riscv/include/asm/sbi.h | 57 +++ arch/sandbox/dts/test.dts | 23 + configs/sandbox_defconfig | 3 + drivers/clk/Kconfig | 7 + drivers/clk/Makefile | 1 + drivers/clk/clk_rpmi.c | 346 +++++++++++++++ drivers/firmware/Kconfig | 1 + drivers/firmware/Makefile | 1 + drivers/firmware/rpmi/Kconfig | 27 ++ drivers/firmware/rpmi/Makefile | 4 + drivers/firmware/rpmi/rpmi-sandbox-clock.c | 446 +++++++++++++++++++ drivers/firmware/rpmi/rpmi-sandbox-power.c | 174 ++++++++ drivers/firmware/rpmi/rpmi-sandbox.c | 158 +++++++ drivers/firmware/rpmi/rpmi-sandbox.h | 48 +++ drivers/firmware/rpmi/rpmi-sbi-mpxy.c | 449 +++++++++++++++++++ drivers/firmware/rpmi/rpmi-shmem.c | 474 +++++++++++++++++++++ drivers/firmware/rpmi/rpmi-uclass.c | 138 ++++++ drivers/power/domain/Kconfig | 7 + drivers/power/domain/Makefile | 1 + drivers/power/domain/rpmi-power-domain.c | 229 ++++++++++ include/dm/uclass-id.h | 1 + include/rpmi-uclass.h | 79 ++++ include/rpmi.h | 159 +++++++ include/rpmi_proto.h | 143 +++++++ test/dm/Makefile | 1 + test/dm/rpmi.c | 69 +++ 27 files changed, 3055 insertions(+) create mode 100644 drivers/clk/clk_rpmi.c create mode 100644 drivers/firmware/rpmi/Kconfig create mode 100644 drivers/firmware/rpmi/Makefile create mode 100644 drivers/firmware/rpmi/rpmi-sandbox-clock.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox-power.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox.h create mode 100644 drivers/firmware/rpmi/rpmi-sbi-mpxy.c create mode 100644 drivers/firmware/rpmi/rpmi-shmem.c create mode 100644 drivers/firmware/rpmi/rpmi-uclass.c create mode 100644 drivers/power/domain/rpmi-power-domain.c create mode 100644 include/rpmi-uclass.h create mode 100644 include/rpmi.h create mode 100644 include/rpmi_proto.h create mode 100644 test/dm/rpmi.c
Signature · 1 line
2.47.3
PATCH 1/7 firmware: add support for RPMI
RISC-V Platform Management Interface (RPMI) defines an OS-agnostic protocol for communication between an Application Processor (AP) and a platform microcontroller (PuC) [1]
Add UCLASS_RPMI to abstract the implementation details of an RPMI transport layer. A transport driver should implement the struct rpmi_transport_ops callback interface while an RPMI client driver should use struct rpmi_chan and the rpmi_* API:
* rpmi_get_by_index()
* rpmi_open()
* rpmi_close()
* rpmi_get_attr()
* rpmi_send_with_resp()
* rpmi_send_posted()
The most important part of the API is rpmi_send_with_resp() and rpmi_send_posted() which is used to send a message and possibly receive a response from the firmware.
rpmi_get_attr() can be used to retrieve attributes of the channel like the implementation ID or the maximum message size. This is to retrieve information that lives outside of the service group of the rpmi_chan, either in the "base service group" or in the transport layer itself.
rpmi_get_by_index() can be used to initialize a struct rpmi_chan using info from the device tree.
rpmi_open() and rpmi_close() must be called once at initialization and deinitialization by clients to let the transport layer do some stateful things if needed.
A helper function is also added: rpmi_check_versions() which is just a wrapper around rpmi_get_attr() to do some basic version checking that all clients should perform.
[1]: https://github.com/riscv-non-isa/riscv-rpmi
--- MAINTAINERS | 6 ++ drivers/firmware/Kconfig | 1 + drivers/firmware/Makefile | 1 + drivers/firmware/rpmi/Kconfig | 9 ++ drivers/firmware/rpmi/Makefile | 1 + drivers/firmware/rpmi/rpmi-uclass.c | 138 ++++++++++++++++++++++++ include/dm/uclass-id.h | 1 + include/rpmi-uclass.h | 79 ++++++++++++++ include/rpmi.h | 159 ++++++++++++++++++++++++++++ include/rpmi_proto.h | 63 +++++++++++ 10 files changed, 458 insertions(+) create mode 100644 drivers/firmware/rpmi/Kconfig create mode 100644 drivers/firmware/rpmi/Makefile create mode 100644 drivers/firmware/rpmi/rpmi-uclass.c create mode 100644 include/rpmi-uclass.h create mode 100644 include/rpmi.h create mode 100644 include/rpmi_proto.h
MAINTAINERS +6 -0
@@ -1689,6 +1689,12 @@ F: drivers/usb/gadget/f_rockusb.c F: cmd/rockusb.c F: doc/README.rockusb +RPMI +M: Charles Perry <[email protected]> +S: Maintained +F: drivers/firmware/rpmi/ +F: include/rpmi* + SANDBOX M: Simon Glass <[email protected]> S: Maintained
drivers/firmware/Kconfig +1 -0
@@ -49,3 +49,4 @@ config ARM_SMCCC_FEATURES source "drivers/firmware/arm-ffa/Kconfig" source "drivers/firmware/scmi/Kconfig" +source "drivers/firmware/rpmi/Kconfig"
drivers/firmware/Makefile +1 -0
@@ -4,3 +4,4 @@ obj-$(CONFIG_TI_SCI_PROTOCOL) += ti_sci.o obj-$(CONFIG_SANDBOX) += firmware-sandbox.o obj-$(CONFIG_ZYNQMP_FIRMWARE) += firmware-zynqmp.o obj-$(CONFIG_SCMI_FIRMWARE) += scmi/ +obj-$(CONFIG_RPMI_FIRMWARE) += rpmi/
drivers/firmware/rpmi/Kconfig new file +9 -0
@@ -0,0 +1,9 @@ +config RPMI_FIRMWARE + bool "Enable RPMI support" + depends on RISCV && DM && OF_CONTROL + select FIRMWARE + help + RISC-V Platform Management Interface (RPMI) defines an OS-agnostic + protocol for communication between an Application Processor (AP) and a + platform microcontroller (PuC). The RPMI specification is available + at: https://github.com/riscv-non-isa/riscv-rpmi
drivers/firmware/rpmi/Makefile new file +1 -0
@@ -0,0 +1 @@ +obj-y += rpmi-uclass.o
drivers/firmware/rpmi/rpmi-uclass.c new file +138 -0
138 lines not on this page · raw message
include/dm/uclass-id.h +1 -0
@@ -128,6 +128,7 @@ enum uclass_id {UCLASS_RESET ,/* Reset controller device */ UCLASS_RKMTD ,/* Rockchip MTD device */ UCLASS_RNG ,/* Random Number Generator */ +UCLASS_RPMI ,/* RPMI transport */ UCLASS_RTC ,/* Real time clock device */ UCLASS_SCMI_AGENT ,/* Interface with an SCMI server */ UCLASS_SCMI_BASE, /* Interface for SCMI Base protocol */
include/rpmi-uclass.h new file +79 -0
79 lines not on this page · raw message
include/rpmi.h new file +159 -0
159 lines not on this page · raw message
include/rpmi_proto.h new file +63 -0
63 lines not on this page · raw message
Signature · 1 line
2.47.3
Re: [PATCH 1/7] firmware: add support for RPMI
Quoted 57 lines · Charles Perry
RISC-V Platform Management Interface (RPMI) defines an OS-agnostic protocol for communication between an Application Processor (AP) and a platform microcontroller (PuC) [1]
Add
UCLASS_RPMIto abstract the implementation details of an RPMI transport layer. A transport driver should implement the structrpmi_transport_opscallback interface while an RPMI client driver should use structrpmi_chanand the rpmi_* API:*
rpmi_get_by_index()
*rpmi_open()
*rpmi_close()
*rpmi_get_attr()
*rpmi_send_with_resp()
*rpmi_send_posted()The most important part of the API is
rpmi_send_with_resp()andrpmi_send_posted()which is used to send a message and possibly receive a response from the firmware.
rpmi_get_attr()can be used to retrieve attributes of the channel like the implementation ID or the maximum message size. This is to retrieve information that lives outside of the service group of therpmi_chan, either in the "base service group" or in the transport layer itself.
rpmi_get_by_index()can be used to initialize a structrpmi_chanusing info from the device tree.
rpmi_open()andrpmi_close()must be called once at initialization and deinitialization by clients to let the transport layer do some stateful things if needed.A helper function is also added:
rpmi_check_versions()which is just a wrapper aroundrpmi_get_attr()to do some basic version checking that all clients should perform.[1]: https://github.com/riscv-non-isa/riscv-rpmi
Signed-off-by Charles Perry <[email protected]>--- MAINTAINERS | 6 ++ drivers/firmware/Kconfig | 1 + drivers/firmware/Makefile | 1 + drivers/firmware/rpmi/Kconfig | 9 ++ drivers/firmware/rpmi/Makefile | 1 + drivers/firmware/rpmi/rpmi-uclass.c | 138 ++++++++++++++++++++++++ include/dm/uclass-id.h | 1 + include/rpmi-uclass.h | 79 ++++++++++++++ include/rpmi.h | 159 ++++++++++++++++++++++++++++ include/rpmi_proto.h | 63 +++++++++++ 10 files changed, 458 insertions(+) create mode 100644 drivers/firmware/rpmi/Kconfig create mode 100644 drivers/firmware/rpmi/Makefile create mode 100644 drivers/firmware/rpmi/rpmi-uclass.c create mode 100644 include/rpmi-uclass.h create mode 100644 include/rpmi.h create mode 100644 include/rpmi_proto.h
...
Quoted 7 lines
drivers/firmware/rpmi/rpmi-uclass.c new file +1
-0@@ -0,0 +1,138 @@ +// SPDX-License-Identifier: GPL-2.0
drivers/firmware/rpmi/rpmi-uclass.c line 11
Quoted 10 lines
drivers/firmware/rpmi/rpmi-uclass.c new file +10
-0+/* +* Copyright (C) 2026 Microchip Technology Inc. All rights reserved. +*/ + +#define LOG_CATEGORY UCLASS_RPMI + +#include <dm.h> +#include <log.h> +#include <rpmi-uclass.h> +#include <dm/device_compat.h>
Should we sort the headers? This applies to rest of the series, too.
...
drivers/firmware/rpmi/rpmi-uclass.c line 109
Quoted 13 lines
drivers/firmware/rpmi/rpmi-uclass.c new file +13
-0+int rpmi_check_versions (struct rpmi_chan *chan,u32 min_proto_version, +u32 servicegroup_id_expect, +u32 min_servicegroup_version) +{ +u32 proto_version, servicegroup_id, servicegroup_version; +int ret; + + ret =rpmi_get_attr (chan,RPMI_ATTR_SPEC_VERSION , &proto_version); +if (ret) +return ret; + +if (proto_version < min_proto_version) { +dev_err (chan->dev,"protocol version mismatch\n" );
The error is caused by a "too old" protocol version instead of a "mismatched" one, shouldn't we be precise here?
Furthermore, it would be good to print the the expected and implemented versions as well to ease debugging. debug() could be used so we don't get the binary size increased in a normal build. So do the servicegroup_id and servicegroup_version's case.
Quoted 24 lines
drivers/firmware/rpmi/rpmi-uclass.c new file +24
-0+return -EINVAL ; + } + + ret =rpmi_get_attr (chan,RPMI_ATTR_SERVICEGROUP_ID , &servicegroup_id); +if (ret) +return ret; + +if (servicegroup_id != servicegroup_id_expect) { +dev_err (chan->dev,"service group match fail\n" ); +return -EINVAL ; + } + + ret =rpmi_get_attr (chan,RPMI_ATTR_SERVICEGROUP_VERSION , + &servicegroup_version); +if (ret) +return ret; + +if (servicegroup_version < min_servicegroup_version) { +dev_err (chan->dev,"service group version mismatch\n" ); +return -EINVAL ; + } + +return 0 ; +}
...
Quoted 5 lines
include/rpmi.h new file +0
-0
...
Quoted 4 lines
include/rpmi.h new file +4
-0+static inline int rpmi_to_linux_error (int rpmi_error) +{ +switch (rpmi_error) { +case RPMI_SUCCESS :
include/rpmi.h line 43
Quoted 10 lines
include/rpmi.h new file +10
-0+return 0 ; +case RPMI_ERR_INVALID_PARAM : +case RPMI_ERR_BAD_RANGE : +case RPMI_ERR_INVALID_STATE : +return -EINVAL ; +case RPMI_ERR_DENIED : +return -EPERM ; +case RPMI_ERR_INVALID_ADDR : +case RPMI_ERR_HW_FAULT : +return -EFAULT ;
RPMI_ERR_HW_FAULT means "failed due to hardware fault" according to RPMI specification, does it really count as "-EFAULT" which is "bad address"?
Quoted 6 lines
include/rpmi.h new file +6
-0+case RPMI_ERR_ALREADY : +return -EALREADY ; +case RPMI_ERR_BUSY : +return -EBUSY ; +case RPMI_ERR_TIMEOUT : +return -ETIMEDOUT ;
include/rpmi.h line 59
Quoted 10 lines
include/rpmi.h new file +10
-0+case RPMI_ERR_IO : +return -ECOMM ; +case RPMI_ERR_FAILED : +case RPMI_ERR_NOTSUPP : +case RPMI_ERR_NO_DATA : +case RPMI_ERR_EXTENSION : +default : +return -EOPNOTSUPP ; + } +}
Regards,
Yao Zi
Re: [PATCH 1/7] firmware: add support for RPMI
Quoted 72 lines · Yao Zi
Quoted 57 lines · Charles Perry
RISC-V Platform Management Interface (RPMI) defines an OS-agnostic protocol for communication between an Application Processor (AP) and a platform microcontroller (PuC) [1]
Add
UCLASS_RPMIto abstract the implementation details of an RPMI transport layer. A transport driver should implement the structrpmi_transport_opscallback interface while an RPMI client driver should use structrpmi_chanand the rpmi_* API:*
rpmi_get_by_index()
*rpmi_open()
*rpmi_close()
*rpmi_get_attr()
*rpmi_send_with_resp()
*rpmi_send_posted()The most important part of the API is
rpmi_send_with_resp()andrpmi_send_posted()which is used to send a message and possibly receive a response from the firmware.
rpmi_get_attr()can be used to retrieve attributes of the channel like the implementation ID or the maximum message size. This is to retrieve information that lives outside of the service group of therpmi_chan, either in the "base service group" or in the transport layer itself.
rpmi_get_by_index()can be used to initialize a structrpmi_chanusing info from the device tree.
rpmi_open()andrpmi_close()must be called once at initialization and deinitialization by clients to let the transport layer do some stateful things if needed.A helper function is also added:
rpmi_check_versions()which is just a wrapper aroundrpmi_get_attr()to do some basic version checking that all clients should perform.[1]: https://github.com/riscv-non-isa/riscv-rpmi
Signed-off-by Charles Perry <[email protected]>--- MAINTAINERS | 6 ++ drivers/firmware/Kconfig | 1 + drivers/firmware/Makefile | 1 + drivers/firmware/rpmi/Kconfig | 9 ++ drivers/firmware/rpmi/Makefile | 1 + drivers/firmware/rpmi/rpmi-uclass.c | 138 ++++++++++++++++++++++++ include/dm/uclass-id.h | 1 + include/rpmi-uclass.h | 79 ++++++++++++++ include/rpmi.h | 159 ++++++++++++++++++++++++++++ include/rpmi_proto.h | 63 +++++++++++ 10 files changed, 458 insertions(+) create mode 100644 drivers/firmware/rpmi/Kconfig create mode 100644 drivers/firmware/rpmi/Makefile create mode 100644 drivers/firmware/rpmi/rpmi-uclass.c create mode 100644 include/rpmi-uclass.h create mode 100644 include/rpmi.h create mode 100644 include/rpmi_proto.h...
Quoted 11 lines
drivers/firmware/rpmi/rpmi-uclass.c new file +5
-0@@ -0,0 +1,138 @@ +// SPDX-License-Identifier: GPL-2.0 +/* +* Copyright (C) 2026 Microchip Technology Inc. All rights reserved. +*/ +
Quoted 8 lines
drivers/firmware/rpmi/rpmi-uclass.c line 11
Quoted 6 lines
drivers/firmware/rpmi/rpmi-uclass.c new file +6
-0+#define LOG_CATEGORY UCLASS_RPMI + +#include <dm.h> +#include <log.h> +#include <rpmi-uclass.h> +#include <dm/device_compat.h> Should we sort the headers? This applies to rest of the series, too.
Ack.
Quoted 20 lines
...
drivers/firmware/rpmi/rpmi-uclass.c line 109
Quoted 13 lines
drivers/firmware/rpmi/rpmi-uclass.c new file +13
-0+int rpmi_check_versions (struct rpmi_chan *chan,u32 min_proto_version, +u32 servicegroup_id_expect, +u32 min_servicegroup_version) +{ +u32 proto_version, servicegroup_id, servicegroup_version; +int ret; + + ret =rpmi_get_attr (chan,RPMI_ATTR_SPEC_VERSION , &proto_version); +if (ret) +return ret; + +if (proto_version < min_proto_version) { +dev_err (chan->dev,"protocol version mismatch\n" );The error is caused by a "too old" protocol version instead of a "mismatched" one, shouldn't we be precise here?
Quoted 4 lines
Furthermore, it would be good to print the the expected and implemented versions as well to ease debugging.
debug()could be used so we don't get the binary size increased in a normal build. So do theservicegroup_idand servicegroup_version's case.
Ack for everything.
Quoted 46 lines
Quoted 25 lines
drivers/firmware/rpmi/rpmi-uclass.c new file +24
-0+return -EINVAL ; + } + + ret =rpmi_get_attr (chan,RPMI_ATTR_SERVICEGROUP_ID , &servicegroup_id); +if (ret) +return ret; + +if (servicegroup_id != servicegroup_id_expect) { +dev_err (chan->dev,"service group match fail\n" ); +return -EINVAL ; + } + + ret =rpmi_get_attr (chan,RPMI_ATTR_SERVICEGROUP_VERSION , + &servicegroup_version); +if (ret) +return ret; + +if (servicegroup_version < min_servicegroup_version) { +dev_err (chan->dev,"service group version mismatch\n" ); +return -EINVAL ; + } + +return 0 ; +}...
Quoted 5 lines
include/rpmi.h new file +0
-0...
Quoted 10 lines
include/rpmi.h new file +10
-0+static inline int rpmi_to_linux_error (int rpmi_error) +{ +switch (rpmi_error) { +case RPMI_SUCCESS : +return 0 ; +case RPMI_ERR_INVALID_PARAM : +case RPMI_ERR_BAD_RANGE : +case RPMI_ERR_INVALID_STATE : +return -EINVAL ; +case RPMI_ERR_DENIED :
Quoted 8 lines
include/rpmi.h line 43
Quoted 4 lines
include/rpmi.h new file +4
-0+return -EPERM ; +case RPMI_ERR_INVALID_ADDR : +case RPMI_ERR_HW_FAULT : +return -EFAULT ;
RPMI_ERR_HW_FAULTmeans "failed due to hardware fault" according to RPMI specification, does it really count as "-EFAULT" which is "bad address"?
This is taken as-is from Linux [1]. So unless there's a good reason to do things differently, I suggest we keep it as-is.
[1]: https://elixir.bootlin.com/linux/v7.1.2/source/include/linux/mailbox/riscv-rpmi-message.h#L62
Thanks,
Charles
PATCH 2/7 firmware: rpmi: add support for the SBI MPXY transport
In reply to Charles PerryThis adds an RPMI transport driver for the Supervisor Binary Interface (SBI) message proxy (MPXY) extension [1].
This is for S-Mode U-Boot where some supervisor software (such as OpenSBI) runs in M-Mode and handles the low level communication with the PuC.
[1]: https://github.com/riscv-non-isa/riscv-sbi-doc (chapter 20)
--- arch/riscv/include/asm/sbi.h | 57 ++++ drivers/firmware/rpmi/Kconfig | 9 + drivers/firmware/rpmi/Makefile | 1 + drivers/firmware/rpmi/rpmi-sbi-mpxy.c | 449 ++++++++++++++++++++++++++ 4 files changed, 516 insertions(+) create mode 100644 drivers/firmware/rpmi/rpmi-sbi-mpxy.c
arch/riscv/include/asm/sbi.h +57 -0
@@ -10,6 +10,7 @@#define _ASM_RISCV_SBI_H#include <linux/types.h> +#include <linux/errno.h> enum sbi_ext_id {SBI_EXT_0_1_SET_TIMER =0x0 ,
@@ -104,6 +105,17 @@ enum sbi_ext_dbcn_fid { SBI_EXT_DBCN_CONSOLE_WRITE_BYTE, }; +enum sbi_ext_mpxy_fid { +SBI_EXT_MPXY_GET_SHMEM_SIZE =0 , +SBI_EXT_MPXY_SET_SHMEM , +SBI_EXT_MPXY_GET_CHANNEL_IDS , +SBI_EXT_MPXY_READ_ATTRS , +SBI_EXT_MPXY_WRITE_ATTRS , +SBI_EXT_MPXY_SEND_MSG_WITH_RESP , +SBI_EXT_MPXY_SEND_MSG_WITHOUT_RESP , +SBI_EXT_MPXY_GET_NOTIFICATION_EVENTS , +}; +#ifdef CONFIG_SBI_V01 #define SBI_EXT_SET_TIMER SBI_EXT_0_1_SET_TIMER#define SBI_FID_SET_TIMER 0
@@ -132,6 +144,14 @@ enum sbi_ext_dbcn_fid {#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f#define SBI_SPEC_VERSION_MINOR_MASK 0xffffff +static inline unsigned long sbi_mk_version (unsigned long major, +unsigned long minor) +{ +return ((major &SBI_SPEC_VERSION_MAJOR_MASK ) + <<SBI_SPEC_VERSION_MAJOR_SHIFT ) | + (minor &SBI_SPEC_VERSION_MINOR_MASK ); +} +/* SBI return error codes */ #define SBI_SUCCESS 0#define SBI_ERR_FAILURE -1
@@ -139,6 +159,43 @@ enum sbi_ext_dbcn_fid {#define SBI_ERR_INVALID_PARAM -3#define SBI_ERR_DENIED -4#define SBI_ERR_INVALID_ADDRESS -5 +#define SBI_ERR_ALREADY_AVAILABLE -6 +#define SBI_ERR_ALREADY_STARTED -7 +#define SBI_ERR_ALREADY_STOPPED -8 +#define SBI_ERR_NO_SHMEM -9 +#define SBI_ERR_INVALID_STATE -10 +#define SBI_ERR_BAD_RANGE -11 +#define SBI_ERR_TIMEOUT -12 +#define SBI_ERR_IO -13 +#define SBI_ERR_DENIED_LOCKED -14 + +static inline int sbi_to_linux_error (int err) +{ +switch (err) { +case SBI_SUCCESS : +return 0 ; +case SBI_ERR_DENIED : +case SBI_ERR_DENIED_LOCKED : +return -EPERM ; +case SBI_ERR_INVALID_PARAM : +case SBI_ERR_INVALID_STATE : +return -EINVAL ; +case SBI_ERR_BAD_RANGE : +return -ERANGE ; +case SBI_ERR_INVALID_ADDRESS : +return -EFAULT ; +case SBI_ERR_NO_SHMEM : +return -ENOMEM ; +case SBI_ERR_TIMEOUT : +return -ETIMEDOUT ; +case SBI_ERR_IO : +return -EIO ; +case SBI_ERR_NOT_SUPPORTED : +case SBI_ERR_FAILURE : +default : +return -EOPNOTSUPP ; + }; +}struct sbiret {long error;
drivers/firmware/rpmi/Kconfig +9 -0
@@ -7,3 +7,12 @@ config RPMI_FIRMWARE protocol for communication between an Application Processor (AP) and a platform microcontroller (PuC). The RPMI specification is available at: https://github.com/riscv-non-isa/riscv-rpmi + +config RPMI_SBI_MPXY + bool "RPMI over SBI MPXY transport" + depends on RPMI_FIRMWARE && SBI + help + Enable support for transport of RPMI messages over the message proxy + (MPXY) extension of the supervisor binary interface (SBI). The + specification for the MPXY extension is available at: + https://github.com/riscv-non-isa/riscv-sbi-doc (chapter 20)
drivers/firmware/rpmi/Makefile +1 -0
@@ -1 +1,2 @@ obj-y += rpmi-uclass.o +obj-$(CONFIG_RPMI_SBI_MPXY) += rpmi-sbi-mpxy.o
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +449 -0
449 lines not on this page · raw message
Signature · 1 line
2.47.3
Re: [PATCH 2/7] firmware: rpmi: add support for the SBI MPXY transport
Quoted 17 lines · Charles Perry
This adds an RPMI transport driver for the Supervisor Binary Interface (SBI) message proxy (MPXY) extension [1].
This is for S-Mode U-Boot where some supervisor software (such as OpenSBI) runs in M-Mode and handles the low level communication with the PuC.
[1]: https://github.com/riscv-non-isa/riscv-sbi-doc (chapter 20)
Signed-off-by Charles Perry <[email protected]>--- arch/riscv/include/asm/sbi.h | 57 ++++ drivers/firmware/rpmi/Kconfig | 9 + drivers/firmware/rpmi/Makefile | 1 + drivers/firmware/rpmi/rpmi-sbi-mpxy.c | 449 ++++++++++++++++++++++++++ 4 files changed, 516 insertions(+) create mode 100644 drivers/firmware/rpmi/rpmi-sbi-mpxy.c
...
Quoted 5 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +0
-0
...
drivers/firmware/rpmi/rpmi-sbi-mpxy.c line 89
Quoted 13 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +13
-0+/* RPMI message protocol specific MPXY attributes */ +enum sbi_mpxy_rpmi_attribute_id { +SBI_MPXY_RPMI_ATTR_SERVICEGROUP_ID =SBI_MPXY_ATTR_MSGPROTO_ATTR_START , +SBI_MPXY_RPMI_ATTR_SERVICEGROUP_VERSION , +SBI_MPXY_RPMI_ATTR_IMPL_ID , +SBI_MPXY_RPMI_ATTR_IMPL_VERSION , +SBI_MPXY_RPMI_ATTR_MAX_ID +}; + +/* +* RPMI specific SBI MPXY channel attributes. +*/ +struct sbi_mpxy_rpmi_channel_attrs {
Would it make sense to organize sbi_mpxy_rpmi_channel_attrs as an array indexed by sbi_mpxy_rpmi_attributed_id subtracted by SBI_MPXY_RPMI_ATTR_MSGPROTO_ATTR_START? With which we could simplify rpmi_sbi_mpxy_get_attr().
Quoted 9 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +9
-0+/* RPMI service group ID */ +u32 servicegroup_id; +/* RPMI service group version */ +u32 servicegroup_version; +/* RPMI implementation ID */ +u32 impl_id; +/* RPMI implementation version */ +u32 impl_version; +};
...
Quoted 31 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +31
-0+static int rpmi_sbi_mpxy_get_attr (struct rpmi_chan *chan, +enum rpmi_attribute_id id,u32 *data) +{ +struct rpmi_sbi_mpxy *mpxy =dev_get_priv (chan->dev); +struct sbi_mpxy_chan_desc *desc = &mpxy->channel_descs[chan->id]; + +switch (id) { +case RPMI_ATTR_SPEC_VERSION : + *data = desc->attrs.msg_proto_version; +break ; +case RPMI_ATTR_MAX_MSG_DATA_SIZE : + *data = desc->max_xfer_len; +break ; +case RPMI_ATTR_SERVICEGROUP_ID : + *data = desc->rpmi_attrs.servicegroup_id; +break ; +case RPMI_ATTR_SERVICEGROUP_VERSION : + *data = desc->rpmi_attrs.servicegroup_version; +break ; +case RPMI_ATTR_IMPL_ID : + *data = desc->rpmi_attrs.impl_id; +break ; +case RPMI_ATTR_IMPL_VERSION : + *data = desc->rpmi_attrs.impl_version; +break ; +default : +return -EOPNOTSUPP ; + } + +return 0 ; +}
...
Quoted 45 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +45
-0+static int rpmi_sbi_mpxy_probe (struct udevice *dev) +{ +struct rpmi_sbi_mpxy *mpxy =dev_get_priv (dev); +struct sbi_mpxy_chan_desc *desc; +unsigned long flags; +struct sbiret sret; +int i, ret; + + mpxy->dev = dev; + +/* Probe for SBI MPXY extension */ +if (sbi_get_spec_version () <sbi_mk_version (1 ,0 ) || +sbi_probe_extension (SBI_EXT_MPXY ) <=0 ) { +dev_info (dev,"SBI MPXY extension not available\n" ); +return -ENODEV ; + } + +/* Find-out shared memory size */ + ret =rpmi_sbi_mpxy_get_shmem_size (&mpxy->shmem_size); +if (ret) { +dev_err (dev,"failed to get MPXY shared memory size\n" ); +return -ENODEV ; + } + + mpxy->shmem =memalign (PAGE_SIZE , mpxy->shmem_size); +if (!mpxy->shmem) +return -ENOMEM ; + +/* Setup shmem in OVERWRITE mode. (flags[1:0] = 00b) */ + flags =0 ; + sret =sbi_ecall (SBI_EXT_MPXY ,SBI_EXT_MPXY_SET_SHMEM , +virt_to_phys (mpxy->shmem),0 , flags,0 ,0 ,0 ); +if (sret.error) { + ret =sbi_to_linux_error (sret.error); + goto err; + } + + ret =rpmi_sbi_mpxy_get_channel_ids (mpxy); +if (ret) + goto err; + +for (i =0 ; i < mpxy->channel_count; i++) { + desc = &mpxy->channel_descs[i]; + desc->max_xfer_len = +min ((u32 )mpxy->shmem_size, desc->attrs.msg_max_len);
drivers/firmware/rpmi/rpmi-sbi-mpxy.c line 411
Quoted 10 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +10
-0+ } + +return 0 ; + +err: +if (mpxy->channel_descs) +free (mpxy->channel_descs); + +if (mpxy->shmem) +free (mpxy->shmem);
free() is no-op when the argument is NULL, so the check is redundant. This applies to rpmi_sbi_mpxy_remove() and other similar cases in the series, too.
Quoted 6 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +6
-0+ +return ret; +} + +static int rpmi_sbi_mpxy_remove (struct udevice *dev) +{
drivers/firmware/rpmi/rpmi-sbi-mpxy.c line 427
Quoted 10 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +10
-0+struct rpmi_sbi_mpxy *mpxy =dev_get_priv (dev); + +if (mpxy->channel_descs) +free (mpxy->channel_descs); + +if (mpxy->shmem) +free (mpxy->shmem); + +return 0 ; +}
Regards,
Yao Zi
Re: [PATCH 2/7] firmware: rpmi: add support for the SBI MPXY transport
Quoted 39 lines · Yao Zi
Quoted 17 lines · Charles Perry
This adds an RPMI transport driver for the Supervisor Binary Interface (SBI) message proxy (MPXY) extension [1].
This is for S-Mode U-Boot where some supervisor software (such as OpenSBI) runs in M-Mode and handles the low level communication with the PuC.
[1]: https://github.com/riscv-non-isa/riscv-sbi-doc (chapter 20)
Signed-off-by Charles Perry <[email protected]>--- arch/riscv/include/asm/sbi.h | 57 ++++ drivers/firmware/rpmi/Kconfig | 9 + drivers/firmware/rpmi/Makefile | 1 + drivers/firmware/rpmi/rpmi-sbi-mpxy.c | 449 ++++++++++++++++++++++++++ 4 files changed, 516 insertions(+) create mode 100644 drivers/firmware/rpmi/rpmi-sbi-mpxy.c...
Quoted 5 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +0
-0...
Quoted 10 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +10
-0+/* RPMI message protocol specific MPXY attributes */ +enum sbi_mpxy_rpmi_attribute_id { +SBI_MPXY_RPMI_ATTR_SERVICEGROUP_ID =SBI_MPXY_ATTR_MSGPROTO_ATTR_START , +SBI_MPXY_RPMI_ATTR_SERVICEGROUP_VERSION , +SBI_MPXY_RPMI_ATTR_IMPL_ID , +SBI_MPXY_RPMI_ATTR_IMPL_VERSION , +SBI_MPXY_RPMI_ATTR_MAX_ID +}; + +/*
Quoted 8 lines
Quoted 3 lines
+ * RPMI specific SBI MPXY channel attributes.
+ */
+structsbi_mpxy_rpmi_channel_attrs{Would it make sense to organize
sbi_mpxy_rpmi_channel_attrsas an array indexed bysbi_mpxy_rpmi_attributed_idsubtracted bySBI_MPXY_RPMI_ATTR_MSGPROTO_ATTR_START? With which we could simplifyrpmi_sbi_mpxy_get_attr().
Yes I'll give this a try. I also find this a bit repetitive that the attributes have an id and a struct where each members are sorted by attribute id. Thanks!
Quoted 98 lines
Quoted 10 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +9
-0+/* RPMI service group ID */ +u32 servicegroup_id; +/* RPMI service group version */ +u32 servicegroup_version; +/* RPMI implementation ID */ +u32 impl_id; +/* RPMI implementation version */ +u32 impl_version; +};...
Quoted 31 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +31
-0+static int rpmi_sbi_mpxy_get_attr (struct rpmi_chan *chan, +enum rpmi_attribute_id id,u32 *data) +{ +struct rpmi_sbi_mpxy *mpxy =dev_get_priv (chan->dev); +struct sbi_mpxy_chan_desc *desc = &mpxy->channel_descs[chan->id]; + +switch (id) { +case RPMI_ATTR_SPEC_VERSION : + *data = desc->attrs.msg_proto_version; +break ; +case RPMI_ATTR_MAX_MSG_DATA_SIZE : + *data = desc->max_xfer_len; +break ; +case RPMI_ATTR_SERVICEGROUP_ID : + *data = desc->rpmi_attrs.servicegroup_id; +break ; +case RPMI_ATTR_SERVICEGROUP_VERSION : + *data = desc->rpmi_attrs.servicegroup_version; +break ; +case RPMI_ATTR_IMPL_ID : + *data = desc->rpmi_attrs.impl_id; +break ; +case RPMI_ATTR_IMPL_VERSION : + *data = desc->rpmi_attrs.impl_version; +break ; +default : +return -EOPNOTSUPP ; + } + +return 0 ; +}...
Quoted 51 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +51
-0+static int rpmi_sbi_mpxy_probe (struct udevice *dev) +{ +struct rpmi_sbi_mpxy *mpxy =dev_get_priv (dev); +struct sbi_mpxy_chan_desc *desc; +unsigned long flags; +struct sbiret sret; +int i, ret; + + mpxy->dev = dev; + +/* Probe for SBI MPXY extension */ +if (sbi_get_spec_version () <sbi_mk_version (1 ,0 ) || +sbi_probe_extension (SBI_EXT_MPXY ) <=0 ) { +dev_info (dev,"SBI MPXY extension not available\n" ); +return -ENODEV ; + } + +/* Find-out shared memory size */ + ret =rpmi_sbi_mpxy_get_shmem_size (&mpxy->shmem_size); +if (ret) { +dev_err (dev,"failed to get MPXY shared memory size\n" ); +return -ENODEV ; + } + + mpxy->shmem =memalign (PAGE_SIZE , mpxy->shmem_size); +if (!mpxy->shmem) +return -ENOMEM ; + +/* Setup shmem in OVERWRITE mode. (flags[1:0] = 00b) */ + flags =0 ; + sret =sbi_ecall (SBI_EXT_MPXY ,SBI_EXT_MPXY_SET_SHMEM , +virt_to_phys (mpxy->shmem),0 , flags,0 ,0 ,0 ); +if (sret.error) { + ret =sbi_to_linux_error (sret.error); + goto err; + } + + ret =rpmi_sbi_mpxy_get_channel_ids (mpxy); +if (ret) + goto err; + +for (i =0 ; i < mpxy->channel_count; i++) { + desc = &mpxy->channel_descs[i]; + desc->max_xfer_len = +min ((u32 )mpxy->shmem_size, desc->attrs.msg_max_len); + } + +return 0 ; + +err: +if (mpxy->channel_descs)
Quoted 8 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c line 411
Quoted 4 lines
drivers/firmware/rpmi/rpmi-sbi-mpxy.c new file +4
-0+free (mpxy->channel_descs); + +if (mpxy->shmem) +free (mpxy->shmem);
free()is no-op when the argument is NULL, so the check is redundant. This applies torpmi_sbi_mpxy_remove()and other similar cases in the series, too.
Ack.
Thanks,
Charles
PATCH 3/7 firmware: rpmi: add support for shared memory transport
In reply to Charles PerryAdd support for transport of RPMI messages over shared memory as described in the RPMI spec [1]. This uses two or four ring buffers composed of a fixed number of "slots" to pass messages between the AP and the PuC. U-Boot is the Application processor (AP) and sends messages through the a2p-req queue and receives response though the p2a-ack queue. There is no support for platform microcontroller (PuC) initiated requests nor notifications.
This transport can be used in M or S mode U-Boot, but in the S-mode case, one must make sure that the supervisor software isn't already using the AP side of the shared memory.
[1]: https://github.com/riscv-non-isa/riscv-rpmi (chapter 2)
--- drivers/firmware/rpmi/Kconfig | 8 + drivers/firmware/rpmi/Makefile | 1 + drivers/firmware/rpmi/rpmi-shmem.c | 474 +++++++++++++++++++++++++++++ include/rpmi_proto.h | 23 ++ 4 files changed, 506 insertions(+) create mode 100644 drivers/firmware/rpmi/rpmi-shmem.c
drivers/firmware/rpmi/Kconfig +8 -0
@@ -16,3 +16,11 @@ config RPMI_SBI_MPXY (MPXY) extension of the supervisor binary interface (SBI). The specification for the MPXY extension is available at: https://github.com/riscv-non-isa/riscv-sbi-doc (chapter 20) + +config RPMI_SHMEM + bool "RPMI over shared memory transport" + depends on RPMI_FIRMWARE + help + Enable support for transport of RPMI messages over shared memory as + described in chapter 2 of the RPMI specification, available at: + https://github.com/riscv-non-isa/riscv-rpmi
drivers/firmware/rpmi/Makefile +1 -0
@@ -1,2 +1,3 @@ obj-y += rpmi-uclass.o obj-$(CONFIG_RPMI_SBI_MPXY) += rpmi-sbi-mpxy.o +obj-$(CONFIG_RPMI_SHMEM) += rpmi-shmem.o
drivers/firmware/rpmi/rpmi-shmem.c new file +474 -0
474 lines not on this page · raw message
include/rpmi_proto.h +23 -0
@@ -60,4 +60,27 @@ enum rpmi_servicegroup_id { RPMI_SRVGRP_VENDOR_END = 0xFFFF, }; +/** RPMI Base ServiceGroup Service IDs */ +enum rpmi_base_service_id { +RPMI_BASE_SRV_ENABLE_NOTIFICATION =0x01 , +RPMI_BASE_SRV_GET_IMPLEMENTATION_VERSION =0x02 , +RPMI_BASE_SRV_GET_IMPLEMENTATION_IDN =0x03 , +RPMI_BASE_SRV_GET_SPEC_VERSION =0x04 , +RPMI_BASE_SRV_GET_PLATFORM_INFO =0x05 , +RPMI_BASE_SRV_PROBE_SERVICE_GROUP =0x06 , +RPMI_BASE_SRV_GET_ATTRIBUTES =0x07 , +}; + +/** RPMI Messages Types */ +enum rpmi_message_type { +/* Normal request backed with ack */ +RPMI_MSG_NORMAL_REQUEST =0x0 , +/* Request without any ack */ +RPMI_MSG_POSTED_REQUEST =0x1 , +/* Acknowledgment for normal request message */ +RPMI_MSG_ACKNOWLDGEMENT =0x2 , +/* Notification message */ +RPMI_MSG_NOTIFICATION =0x3 , +}; +#endif
Signature · 1 line
2.47.3
PATCH 4/7 drivers: clk: add support for RPMI clocks
In reply to Charles PerryThe RISC-V Platform Management Interface (RPMI) defines a service group for control and monitoring of clocks [1]. This can be exposed as a UCLASS_CLK driver.
[1]: https://github.com/riscv-non-isa/riscv-rpmi (chapter 4.8)
--- MAINTAINERS | 1 + drivers/clk/Kconfig | 7 + drivers/clk/Makefile | 1 + drivers/clk/clk_rpmi.c | 346 +++++++++++++++++++++++++++++++++++++++++ include/rpmi_proto.h | 39 +++++ 5 files changed, 394 insertions(+) create mode 100644 drivers/clk/clk_rpmi.c
MAINTAINERS +1 -0
@@ -1692,6 +1692,7 @@ F: doc/README.rockusb RPMI M: Charles Perry <[email protected]> S: Maintained +F: drivers/clk/clk_rpmi.c F: drivers/firmware/rpmi/ F: include/rpmi*
drivers/clk/Kconfig +7 -0
@@ -181,6 +181,13 @@ config CLK_OCTEON help Enable this to support the clocks on Octeon MIPS platforms. +config CLK_RPMI + bool "Enable RPMI clock driver" + depends on CLK && RPMI_FIRMWARE + help + Enable this option if you want to support clock devices exposed + by RISC-V Platform Management Interface (RPMI). + config SANDBOX_CLK_CCF bool "Sandbox Common Clock Framework [CCF] support" depends on SANDBOX
drivers/clk/Makefile +1 -0
@@ -45,6 +45,7 @@ obj-$(CONFIG_CLK_OCTEON) += clk_octeon.o obj-$(CONFIG_CLK_OWL) += owl/ obj-$(CONFIG_CLK_QCOM) += qcom/ obj-$(CONFIG_CLK_RENESAS) += renesas/ +obj-$(CONFIG_CLK_RPMI) += clk_rpmi.o obj-$(CONFIG_$(PHASE_)CLK_SCMI) += clk_scmi.o obj-$(CONFIG_CLK_SIFIVE) += sifive/ obj-$(CONFIG_CLK_SOPHGO) += sophgo/
drivers/clk/clk_rpmi.c new file +346 -0
346 lines not on this page · raw message
include/rpmi_proto.h +39 -0
@@ -83,4 +83,43 @@ enum rpmi_message_type { RPMI_MSG_NOTIFICATION = 0x3, }; +/* RPMI clock service IDs */ +enum rpmi_clock_service_id { +RPMI_CLK_SRV_ENABLE_NOTIFICATION =0x01 , +RPMI_CLK_SRV_GET_NUM_CLOCKS =0x02 , +RPMI_CLK_SRV_GET_ATTRIBUTES =0x03 , +RPMI_CLK_SRV_GET_SUPPORTED_RATES =0x04 , +RPMI_CLK_SRV_SET_CONFIG =0x05 , +RPMI_CLK_SRV_GET_CONFIG =0x06 , +RPMI_CLK_SRV_SET_RATE =0x07 , +RPMI_CLK_SRV_GET_RATE =0x08 , +RPMI_CLK_SRV_ID_MAX_COUNT +}; + +/** Clock rate match mode */ +enum rpmi_clock_rate_match { +RPMI_CLK_RATE_MATCH_PLATFORM =0 , +RPMI_CLK_RATE_MATCH_ROUND_DOWN =1 , +RPMI_CLK_RATE_MATCH_ROUND_UP =2 , +RPMI_CLK_RATE_MATCH_MAX +}; + +/** Supported clock states */ +enum rpmi_clock_state { +RPMI_CLK_STATE_DISABLED =0 , +RPMI_CLK_STATE_ENABLED =1 , +RPMI_CLK_STATE_MAX +}; + +/** Clock type based on rate format */ +enum rpmi_clock_type { +RPMI_CLK_TYPE_DISCRETE =0 , +RPMI_CLK_TYPE_LINEAR =1 , +RPMI_CLK_TYPE_MAX +}; + +#define RPMI_CLK_NAME_LEN 16 + +#define RPMI_CLOCK_RATE_INVALID (-1ULL) +#endif
Signature · 1 line
2.47.3
Re: [PATCH 4/7] drivers: clk: add support for RPMI clocks
Quoted 15 lines · Charles Perry
The RISC-V Platform Management Interface (RPMI) defines a service group for control and monitoring of clocks [1]. This can be exposed as a
UCLASS_CLKdriver.[1]: https://github.com/riscv-non-isa/riscv-rpmi (chapter 4.8)
Signed-off-by Charles Perry <[email protected]>--- MAINTAINERS | 1 + drivers/clk/Kconfig | 7 + drivers/clk/Makefile | 1 + drivers/clk/clk_rpmi.c | 346 +++++++++++++++++++++++++++++++++++++++++ include/rpmi_proto.h | 39 +++++ 5 files changed, 394 insertions(+) create mode 100644 drivers/clk/clk_rpmi.c
...
Quoted 6 lines
drivers/clk/clk_rpmi.c new file +0
-0@@ -0,0 +1,346 @@
...
drivers/clk/clk_rpmi.c line 55
Quoted 3 lines
drivers/clk/clk_rpmi.c new file +3
-0+struct rpmi_clk_def { +u32 num_rates; +u32 transition_latency;
rpmi_clk_def.transition_latency only gets assigned in rpmi_clk_get_attrs() and never used. Since U-Boot's clock infrastructure doesn't care about it at all, maybe we could drop the field?
Quoted 3 lines
drivers/clk/clk_rpmi.c new file +3
-0+enum rpmi_clock_type type; +char name[RPMI_CLK_NAME_LEN +1 ]; +};
...
Quoted 11 lines
drivers/clk/clk_rpmi.c new file +11
-0+static int _rpmi_clk_enable_disable (struct rpmi_clk_priv *priv,u32 clkid, +bool ena) +{ +struct rpmi_set_config_tx tx = { + .clkid =cpu_to_le32 (clkid), + .config =cpu_to_le32 (ena ?RPMI_CLK_STATE_ENABLED : +RPMI_CLK_STATE_DISABLED ), + }; +__le32 rx; +int ret, status; +
drivers/clk/clk_rpmi.c line 86
Quoted 10 lines
drivers/clk/clk_rpmi.c new file +10
-0+ ret =rpmi_send_with_resp (&priv->chan,RPMI_CLK_SRV_SET_CONFIG , &tx, +sizeof (tx), &rx,sizeof (rx),NULL ); +if (ret) +return ret; + + status =le32_to_cpu (rx); +if (status) +return rpmi_to_linux_error (status); + +return 0 ;
Please return rpmi_to_linux_error(status) and drop the if above.
Quoted 19 lines
drivers/clk/clk_rpmi.c new file +19
-0+} + +static ulong _rpmi_clk_get_rate (struct rpmi_clk_priv *priv,u32 clkid) +{ +__le32 tx =cpu_to_le32 (clkid); +struct rpmi_get_rate_rx rx; +int ret, status; + + ret =rpmi_send_with_resp (&priv->chan,RPMI_CLK_SRV_GET_RATE , &tx, +sizeof (tx), &rx,sizeof (rx),NULL ); +if (ret) +return ret; + + status =le32_to_cpu (rx.status); +if (status) +return rpmi_to_linux_error (status); + +return (((u64)(le32_to_cpu (rx.hi)) <<32 ) | (u32)(le32_to_cpu (rx.lo))); +}
...
drivers/clk/clk_rpmi.c line 127
Quoted 2 lines
drivers/clk/clk_rpmi.c new file +2
-0+static int _rpmi_clk_set_rate (struct rpmi_clk_priv *priv,u32 clkid, +unsigned long rate)
This function is declared to return int, but...
Quoted 10 lines
drivers/clk/clk_rpmi.c new file +10
-0+{ +struct rpmi_set_rate_tx tx = { + .clkid =cpu_to_le32 (clkid), + .flags =0 , + .lo =cpu_to_le32 (lower_32_bits (rate)), + .hi =cpu_to_le32 (upper_32_bits (rate)), + }; +__le32 rx; +int ret, status; +
drivers/clk/clk_rpmi.c line 147
Quoted 10 lines
drivers/clk/clk_rpmi.c new file +10
-0+ ret =rpmi_send_with_resp (&priv->chan,RPMI_CLK_SRV_SET_RATE , &tx, +sizeof (tx), &rx,sizeof (rx),NULL ); +if (ret) +return ret; + + status =le32_to_cpu (rx); +if (status) +return rpmi_to_linux_error (status); + +return _rpmi_clk_get_rate (priv, clkid);
_rpmi_clk_get_rate() returns ulong. This unnecessarily truncates the rate on 64bit platform when it's above approximately 2.1GHz.
Quoted 1 line
+}
Regards,
Yao Zi
Re: [PATCH 4/7] drivers: clk: add support for RPMI clocks
Quoted 28 lines · Yao Zi
Quoted 15 lines · Charles Perry
The RISC-V Platform Management Interface (RPMI) defines a service group for control and monitoring of clocks [1]. This can be exposed as a
UCLASS_CLKdriver.[1]: https://github.com/riscv-non-isa/riscv-rpmi (chapter 4.8)
Signed-off-by Charles Perry <[email protected]>--- MAINTAINERS | 1 + drivers/clk/Kconfig | 7 + drivers/clk/Makefile | 1 + drivers/clk/clk_rpmi.c | 346 +++++++++++++++++++++++++++++++++++++++++ include/rpmi_proto.h | 39 +++++ 5 files changed, 394 insertions(+) create mode 100644 drivers/clk/clk_rpmi.c...
Quoted 6 lines
drivers/clk/clk_rpmi.c new file +0
-0@@ -0,0 +1,346 @@...
Quoted 7 lines
drivers/clk/clk_rpmi.c line 55
Quoted 3 lines
drivers/clk/clk_rpmi.c new file +3
-0+struct rpmi_clk_def { +u32 num_rates; +u32 transition_latency;rpmi_clk_def.transition_latency only gets assigned in
rpmi_clk_get_attrs()and never used. Since U-Boot's clock infrastructure doesn't care about it at all, maybe we could drop the field?
Ack.
Quoted 22 lines
Quoted 4 lines
drivers/clk/clk_rpmi.c new file +3
-0+enum rpmi_clock_type type; +char name[RPMI_CLK_NAME_LEN +1 ]; +};...
Quoted 15 lines
drivers/clk/clk_rpmi.c new file +15
-0+static int _rpmi_clk_enable_disable (struct rpmi_clk_priv *priv,u32 clkid, +bool ena) +{ +struct rpmi_set_config_tx tx = { + .clkid =cpu_to_le32 (clkid), + .config =cpu_to_le32 (ena ?RPMI_CLK_STATE_ENABLED : +RPMI_CLK_STATE_DISABLED ), + }; +__le32 rx; +int ret, status; + + ret =rpmi_send_with_resp (&priv->chan,RPMI_CLK_SRV_SET_CONFIG , &tx, +sizeof (tx), &rx,sizeof (rx),NULL ); +if (ret) +return ret;
Quoted 8 lines
drivers/clk/clk_rpmi.c line 86
Quoted 6 lines
drivers/clk/clk_rpmi.c new file +6
-0+ + status =le32_to_cpu (rx); +if (status) +return rpmi_to_linux_error (status); + +return 0 ;Please return
rpmi_to_linux_error(status)and drop the if above.
Ack.
Quoted 43 lines
Quoted 20 lines
drivers/clk/clk_rpmi.c new file +19
-0+} + +static ulong _rpmi_clk_get_rate (struct rpmi_clk_priv *priv,u32 clkid) +{ +__le32 tx =cpu_to_le32 (clkid); +struct rpmi_get_rate_rx rx; +int ret, status; + + ret =rpmi_send_with_resp (&priv->chan,RPMI_CLK_SRV_GET_RATE , &tx, +sizeof (tx), &rx,sizeof (rx),NULL ); +if (ret) +return ret; + + status =le32_to_cpu (rx.status); +if (status) +return rpmi_to_linux_error (status); + +return (((u64)(le32_to_cpu (rx.hi)) <<32 ) | (u32)(le32_to_cpu (rx.lo))); +}...
drivers/clk/clk_rpmi.c line 127
Quoted 2 lines
drivers/clk/clk_rpmi.c new file +2
-0+static int _rpmi_clk_set_rate (struct rpmi_clk_priv *priv,u32 clkid, +unsigned long rate)This function is declared to return int, but...
Quoted 15 lines
drivers/clk/clk_rpmi.c new file +15
-0+{ +struct rpmi_set_rate_tx tx = { + .clkid =cpu_to_le32 (clkid), + .flags =0 , + .lo =cpu_to_le32 (lower_32_bits (rate)), + .hi =cpu_to_le32 (upper_32_bits (rate)), + }; +__le32 rx; +int ret, status; + + ret =rpmi_send_with_resp (&priv->chan,RPMI_CLK_SRV_SET_RATE , &tx, +sizeof (tx), &rx,sizeof (rx),NULL ); +if (ret) +return ret; +
Quoted 8 lines
drivers/clk/clk_rpmi.c line 147
Quoted 5 lines
drivers/clk/clk_rpmi.c new file +5
-0+ status =le32_to_cpu (rx); +if (status) +return rpmi_to_linux_error (status); + +return _rpmi_clk_get_rate (priv, clkid);
_rpmi_clk_get_rate()returns ulong. This unnecessarily truncates the rate on 64bit platform when it's above approximately 2.1GHz.
Good catch.
I'll keep the return code as int here and put the call to _rpmi_clk_get_rate() inside rpmi_clk_set_rate() instead.
Thanks,
Charles
PATCH 5/7 drivers: power: add support for RPMI power domains
In reply to Charles PerryThe RISC-V Platform Management Interface (RPMI) defines the device power service group [1] that is used to control power domains. This is used to implement a UCLASS_POWER_DOMAIN driver.
[1]: https://github.com/riscv-non-isa/riscv-rpmi (chapter 4.9)
--- MAINTAINERS | 1 + drivers/power/domain/Kconfig | 7 + drivers/power/domain/Makefile | 1 + drivers/power/domain/rpmi-power-domain.c | 229 +++++++++++++++++++++++ include/rpmi_proto.h | 18 ++ 5 files changed, 256 insertions(+) create mode 100644 drivers/power/domain/rpmi-power-domain.c
MAINTAINERS +1 -0
@@ -1694,6 +1694,7 @@ M: Charles Perry <[email protected]> S: Maintained F: drivers/clk/clk_rpmi.c F: drivers/firmware/rpmi/ +F: drivers/power/domain/rpmi-power-domain.c F: include/rpmi* SANDBOX
drivers/power/domain/Kconfig +7 -0
@@ -106,6 +106,13 @@ config RENESAS_R8A78000_POWER_DOMAIN and reset driver. The MDLC is responsible for managing both power domains and resets on R-Car R8A78000 X5H SoC. +config RPMI_POWER_DOMAIN + bool "Enable the RPMI Power domain driver" + depends on POWER_DOMAIN && RPMI_FIRMWARE + help + Enable this option if you want to support power domains managed by + the RISC-V Platform Management Interface (RPMI) protocol. + config SANDBOX_POWER_DOMAIN bool "Enable the sandbox power domain test driver" depends on POWER_DOMAIN && SANDBOX
drivers/power/domain/Makefile +1 -0
@@ -16,6 +16,7 @@ obj-$(CONFIG_MESON_GX_VPU_POWER_DOMAIN) += meson-gx-pwrc-vpu.o obj-$(CONFIG_MESON_EE_POWER_DOMAIN) += meson-ee-pwrc.o obj-$(CONFIG_MESON_SECURE_POWER_DOMAIN) += meson-secure-pwrc.o obj-$(CONFIG_RENESAS_R8A78000_POWER_DOMAIN) += renesas-r8a78000-power-domain.o +obj-$(CONFIG_RPMI_POWER_DOMAIN) += rpmi-power-domain.o obj-$(CONFIG_SANDBOX_POWER_DOMAIN) += sandbox-power-domain.o obj-$(CONFIG_SANDBOX_POWER_DOMAIN) += sandbox-power-domain-test.o obj-$(CONFIG_SCMI_POWER_DOMAIN) += scmi-power-domain.o
drivers/power/domain/rpmi-power-domain.c new file +229 -0
229 lines not on this page · raw message
include/rpmi_proto.h +18 -0
@@ -122,4 +122,22 @@ enum rpmi_clock_type {#define RPMI_CLOCK_RATE_INVALID (-1ULL) +/* RPMI Device_Power (DPWR) service IDs */ +enum rpmi_dpwr_service_id { +RPMI_DPWR_SRV_ENABLE_NOTIFICATION =0x01 , +RPMI_DPWR_SRV_GET_NUM_DOMAINS =0x02 , +RPMI_DPWR_SRV_GET_ATTRIBUTES =0x03 , +RPMI_DPWR_SRV_SET_DPWR_STATE =0x04 , +RPMI_DPWR_SRV_GET_DPWR_STATE =0x05 , +RPMI_DPWR_SRV_ID_MAX , +}; + +enum rpmi_dpwr_state { +RPMI_DPWR_STATE_INVALID =-1 , +RPMI_DPWR_STATE_ON =0 , +RPMI_DPWR_STATE_OFF =3 , +}; + +#define RPMI_DPWR_NAME_LEN 16 +#endif
Signature · 1 line
2.47.3
Re: [PATCH 5/7] drivers: power: add support for RPMI power domains
Quoted 8 lines · Charles Perry
The RISC-V Platform Management Interface (RPMI) defines the device power service group [1] that is used to control power domains. This is used to implement a
UCLASS_POWER_DOMAINdriver.[1]: https://github.com/riscv-non-isa/riscv-rpmi (chapter 4.9)
Signed-off-by Charles Perry <[email protected]>---
...
drivers/power/domain/rpmi-power-domain.c line 215
Quoted 4 lines
drivers/power/domain/rpmi-power-domain.c new file +4
-0+}; + +static const struct udevice_id rpmi_pd_ids[] = { + { .compatible ="riscv,rpmi-device-power" },
Not find this binding in Linux kernel or dt-schema.
Regards
Peng.
Re: [PATCH 5/7] drivers: power: add support for RPMI power domains
Quoted 17 lines · Peng Fan
Quoted 8 lines · Charles Perry
The RISC-V Platform Management Interface (RPMI) defines the device power service group [1] that is used to control power domains. This is used to implement a
UCLASS_POWER_DOMAINdriver.[1]: https://github.com/riscv-non-isa/riscv-rpmi (chapter 4.9)
Signed-off-by Charles Perry <[email protected]>---
...
drivers/power/domain/rpmi-power-domain.c line 215
Quoted 4 lines
drivers/power/domain/rpmi-power-domain.c new file +4
-0+}; + +static const struct udevice_id rpmi_pd_ids[] = { + { .compatible ="riscv,rpmi-device-power" },Not find this binding in Linux kernel or dt-schema.
Quoted 2 lines
Regards
Peng.
Right.
I'll remove the device power service group in the next version. This is not strictly needed for the initial introduction of RPMI and can be added latter.
Thanks,
Charles
PATCH 6/7 firmware: rpmi: add a test and a sandbox driver
In reply to Charles PerryCreate a new transport layer that sends the messages to a simulated RPMI firmware.
For now, this only contains a simulation of the clock service group which is implemented using code from librpmi [1] and adapted.
Also add a test in test/dm that makes use of the RPMI clock driver.
--- MAINTAINERS | 1 + arch/sandbox/dts/test.dts | 12 + configs/sandbox_defconfig | 2 + drivers/firmware/rpmi/Kconfig | 3 +- drivers/firmware/rpmi/Makefile | 1 + drivers/firmware/rpmi/rpmi-sandbox-clock.c | 446 +++++++++++++++++++++ drivers/firmware/rpmi/rpmi-sandbox.c | 156 +++++++ drivers/firmware/rpmi/rpmi-sandbox.h | 47 +++ test/dm/Makefile | 1 + test/dm/rpmi.c | 48 +++ 10 files changed, 716 insertions(+), 1 deletion(-) create mode 100644 drivers/firmware/rpmi/rpmi-sandbox-clock.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox.h create mode 100644 test/dm/rpmi.c
MAINTAINERS +1 -0
@@ -1696,6 +1696,7 @@ F: drivers/clk/clk_rpmi.c F: drivers/firmware/rpmi/ F: drivers/power/domain/rpmi-power-domain.c F: include/rpmi* +F: test/dm/rpmi.c SANDBOX M: Simon Glass <[email protected]>
arch/sandbox/dts/test.dts +12 -0
@@ -874,6 +874,18 @@ sm: secure-monitor { compatible = "sandbox,sm"; }; + + rpmi: rpmi { + compatible = "sandbox,rpmi-transport"; + #mbox-cells = <1>; + }; + + rpmi_clock: rpmi-clock { + compatible = "riscv,rpmi-clock"; + mboxes = <&rpmi 0x8>; + #clock-cells = <1>; + clocks = <&rpmi_clock 0>, <&rpmi_clock 1>, <&rpmi_clock 2>, <&rpmi_clock 3>; + }; }; fpga {
configs/sandbox_defconfig +2 -0
@@ -191,6 +191,7 @@ CONFIG_BUTTON_GPIO=y CONFIG_CLK=y CONFIG_CLK_K210=y CONFIG_CLK_K210_SET_RATE=y +CONFIG_CLK_RPMI=y CONFIG_SANDBOX_CLK_CCF=y CONFIG_CLK_SCMI=y CONFIG_CPU=y
@@ -207,6 +208,7 @@ CONFIG_FASTBOOT_FLASH=y CONFIG_FASTBOOT_FLASH_MMC_DEV=0 CONFIG_ARM_FFA_TRANSPORT=y CONFIG_SCMI_FIRMWARE=y +CONFIG_RPMI_FIRMWARE=y CONFIG_FPGA_ALTERA=y CONFIG_FPGA_STRATIX_II=y CONFIG_FPGA_STRATIX_V=y
drivers/firmware/rpmi/Kconfig +2 -1
@@ -1,6 +1,7 @@ config RPMI_FIRMWARE bool "Enable RPMI support" - depends on RISCV && DM && OF_CONTROL + depends on RISCV || SANDBOX + depends on DM && OF_CONTROL select FIRMWARE help RISC-V Platform Management Interface (RPMI) defines an OS-agnostic
drivers/firmware/rpmi/Makefile +1 -0
@@ -1,3 +1,4 @@ obj-y += rpmi-uclass.o obj-$(CONFIG_RPMI_SBI_MPXY) += rpmi-sbi-mpxy.o obj-$(CONFIG_RPMI_SHMEM) += rpmi-shmem.o +obj-$(CONFIG_SANDBOX) += rpmi-sandbox.o rpmi-sandbox-clock.o
drivers/firmware/rpmi/rpmi-sandbox-clock.c new file +446 -0
446 lines not on this page · raw message
drivers/firmware/rpmi/rpmi-sandbox.c new file +156 -0
156 lines not on this page · raw message
drivers/firmware/rpmi/rpmi-sandbox.h new file +47 -0
47 lines not on this page · raw message
test/dm/Makefile +1 -0
@@ -109,6 +109,7 @@ obj-$(CONFIG_CMD_RKMTD) += rkmtd.o obj-$(CONFIG_$(PHASE_)DM_RNG) += rng.o obj-$(CONFIG_DM_RTC) += rtc.o obj-$(CONFIG_SCMI_FIRMWARE) += scmi.o +obj-$(CONFIG_RPMI_FIRMWARE) += rpmi.o obj-$(CONFIG_SCSI) += scsi.o obj-$(CONFIG_DM_SERIAL) += serial.o obj-$(CONFIG_DM_SPI_FLASH) += sf.o
test/dm/rpmi.c new file +48 -0
48 lines not on this page · raw message
Signature · 1 line
2.47.3
PATCH 7/7 firmware: rpmi: add a test and sandbox for device power
In reply to Charles PerryAdd a simulation of the device power service group using code from librpmi [1] and adapted.
--- arch/sandbox/dts/test.dts | 11 ++ configs/sandbox_defconfig | 1 + drivers/firmware/rpmi/Makefile | 2 +- drivers/firmware/rpmi/rpmi-sandbox-power.c | 174 +++++++++++++++++++++ drivers/firmware/rpmi/rpmi-sandbox.c | 2 + drivers/firmware/rpmi/rpmi-sandbox.h | 1 + test/dm/rpmi.c | 21 +++ 7 files changed, 211 insertions(+), 1 deletion(-) create mode 100644 drivers/firmware/rpmi/rpmi-sandbox-power.c
arch/sandbox/dts/test.dts +11 -0
@@ -886,6 +886,12 @@ #clock-cells = <1>; clocks = <&rpmi_clock 0>, <&rpmi_clock 1>, <&rpmi_clock 2>, <&rpmi_clock 3>; }; + + rpmi_dpwr: rpmi-power-domain { + compatible = "riscv,rpmi-device-power"; + mboxes = <&rpmi 0x9>; + #power-domain-cells = <1>; + }; }; fpga {
@@ -1927,6 +1933,11 @@ regul1-supply = <®ul1_scmi>; }; + rpmi-power-domain-test { + compatible = "sandbox,power-domain-test"; + power-domains = <&rpmi_dpwr 1>; + }; + pinctrl { compatible = "sandbox,pinctrl";
configs/sandbox_defconfig +1 -0
@@ -289,6 +289,7 @@ CONFIG_PINCONF=y CONFIG_PINCTRL_SANDBOX=y CONFIG_PINCTRL_SINGLE=y CONFIG_POWER_DOMAIN=y +CONFIG_RPMI_POWER_DOMAIN=y CONFIG_SANDBOX_POWER_DOMAIN=y CONFIG_SCMI_POWER_DOMAIN=y CONFIG_DM_PMIC=y
drivers/firmware/rpmi/Makefile +1 -1
@@ -1,4 +1,4 @@ obj-y += rpmi-uclass.o obj-$(CONFIG_RPMI_SBI_MPXY) += rpmi-sbi-mpxy.o obj-$(CONFIG_RPMI_SHMEM) += rpmi-shmem.o -obj-$(CONFIG_SANDBOX) += rpmi-sandbox.o rpmi-sandbox-clock.o +obj-$(CONFIG_SANDBOX) += rpmi-sandbox.o rpmi-sandbox-clock.o rpmi-sandbox-power.o
drivers/firmware/rpmi/rpmi-sandbox-power.c new file +174 -0
174 lines not on this page · raw message
drivers/firmware/rpmi/rpmi-sandbox.c +2 -0
@@ -29,6 +29,8 @@ const struct rpmi_service_group *rpmi_sandbox_sg(u32 servicegroup_id)switch (servicegroup_id) {case RPMI_SRVGRP_CLOCK :return &rpmi_sandbox_clock_sg; +case RPMI_SRVGRP_DEVICE_POWER : +return &rpmi_sandbox_dpwr_sg;default :return NULL ; }
drivers/firmware/rpmi/rpmi-sandbox.h +1 -0
@@ -43,5 +43,6 @@ struct rpmi_service_group { };extern const struct rpmi_service_group rpmi_sandbox_clock_sg; +extern const struct rpmi_service_group rpmi_sandbox_dpwr_sg;#endif
test/dm/rpmi.c +21 -0
@@ -9,6 +9,7 @@#include <stdio.h> #include <dm/device-internal.h> #include <dm/test.h> +#include <asm/power-domain.h> #include <test/ut.h> static int dm_test_rpmi_clock (struct unit_test_state *uts)
@@ -46,3 +47,23 @@ static int dm_test_rpmi_clock(struct unit_test_state *uts)return 0 ; }DM_TEST (dm_test_rpmi_clock,UTF_SCAN_FDT ); + +static int dm_test_rpmi_dpwr (struct unit_test_state *uts) +{ +struct udevice *rpmi_dev, *dpwr_dev, *dpwr_consumer; + +ut_assertok (uclass_get_device_by_name (UCLASS_RPMI ,"rpmi" , &rpmi_dev)); + +ut_assertok (uclass_get_device_by_name (UCLASS_POWER_DOMAIN , +"rpmi-power-domain" , &dpwr_dev)); + +ut_assertok (uclass_get_device_by_name ( +UCLASS_MISC ,"rpmi-power-domain-test" , &dpwr_consumer)); + +ut_assertok (sandbox_power_domain_test_get (dpwr_consumer)); +ut_assertok (sandbox_power_domain_test_on (dpwr_consumer)); +ut_assertok (sandbox_power_domain_test_off (dpwr_consumer)); + +return 0 ; +} +DM_TEST (dm_test_rpmi_dpwr,UTF_SCAN_FDT );
Signature · 1 line
2.47.3
Quoted 100 lines · Charles Perry
Hello,
This series adds support for RISC-V Platform Management Interface (RPMI) to U-Boot. RPMI is an OS-agnostic protocol for communication between an Application Processor (AP) and a Platform Microcontroller (PuC) [1]. The goals and purpose of RPMI are similar to ARM's SCMI.
This patchset first adds an abstraction layer for the transport (
UCLASS_RPMI), then adds two transport layer (SBI MPXY and shared memory) and finally adds two client, or service group in RPMI jargon: Clock and device power service groups.This series also include a sandbox mode for which the RPMI transport doesn't attach to a firmware but contains a simulation of a firmware. This reuses code from librpmi [2] to create the message responses.
Apart from the spec [1], this series reuses code from the Linux SBI MPXY driver [3], OpenSBI shared memory transport [4] and Linux RPMI clock driver [5].
This was tested on a PIC64-HPSC SoC which features an application cluster and a system controller with isolated address spaces, hence the need for RPMI. I've tested the interoperability between RPMI clients and transports with an S-Mode U-boot (using the MPXY transport) and M-Mode U-Boot (using shared memory transport) and the following device tree fragment:
``` firmware { rpmi: mailbox@7000000000 { compatible = "riscv,rpmi-shmem-mbox"; reg = <0x70 0x00000000 0x0 0x3000>, <0x70 0x00003000 0x0 0x3000>, <0x70 0x00006000 0x0 0x1000>, <0x70 0x00007000 0x0 0x1000>; reg-names = "a2p-req", "p2a-ack", "p2a-req", "a2p-ack"; riscv,slot-size = <64>; #mbox-cells = <1>; }; ifndef CONFIG_RISCV_MMODE clock-service { compatible = "riscv,rpmi-mpxy-clock"; mboxes = <&rpmi 0x8>; riscv,sbi-mpxy-channel-id = <0x1000>; }; device-power-service { compatible = "riscv,rpmi-mpxy-device-power"; mboxes = <&rpmi 0x9>; riscv,sbi-mpxy-channel-id = <0x1001>; }; mpxy_mbox: mailbox { compatible = "riscv,sbi-mpxy-mbox"; #mbox-cells = <2>; }; endif sysc_clk: clock-controller { compatible = "riscv,rpmi-clock"; ifdef CONFIG_RISCV_MMODE mboxes = <&rpmi 0x8>; else mboxes = <&mpxy_mbox 0x1000 0x0>; endif #clock-cells = <1>; }; sysc_dpwr: power-domain-controller { compatible = "riscv,rpmi-device-power"; ifdef CONFIG_RISCV_MMODE mboxes = <&rpmi 0x9>; else mboxes = <&mpxy_mbox 0x1001 0x0>; endif #power-domain-cells = <1>; }; }; ```[1]: https://github.com/riscv-non-isa/riscv-rpmi
[2]: https://github.com/riscv-software-src/librpmi
[3]: https://elixir.bootlin.com/linux/v7.0.10/source/drivers/mailbox/riscv-sbi-mpxy-mbox.c
[4]: https://elixir.bootlin.com/opensbi/v1.8.1/source/lib/utils/mailbox/fdt_mailbox_rpmi_shmem.c
[5]: https://elixir.bootlin.com/linux/v7.0.10/source/drivers/clk/clk-rpmi.cCc Rahul Pathak <[email protected]>Cc Anup Patel <[email protected]>Cc Paul Walmsley <[email protected]>Cc Palmer Dabbelt <[email protected]>Cc Albert Ou <[email protected]>Cc Alexandre Ghiti <[email protected]>Cc Romain Caritey <[email protected]>Cc Mame Maria Mbaye <[email protected]>Cc Tom Rini <[email protected]>Cc Heinrich Schuchardt <[email protected]>Cc Rick Chen <[email protected]>Cc Leo Yu-Chi Liang <[email protected]>
Quoted 8 lines
Charles Perry (7):
firmware: add support for RPMI
firmware: rpmi: add support for the SBI MPXY transport
firmware: rpmi: add support for shared memory transport
drivers: clk: add support for RPMI clocks
drivers: power: add support for RPMI power domains
firmware: rpmi: add a test and a sandbox driver
firmware: rpmi: add a test and sandbox for device power
I have a more general question.
Why add separate firmware framework for RPMI ?
You can just extend the existing U-Boot mailbox framework and just like we have in Linux. The separate transport drivers will be mailbox-controller drivers and each service group will have a separate mailbox-client driver under appropriate driver subsystem.
Quoted 45 lines
MAINTAINERS | 9 + arch/riscv/include/asm/sbi.h | 57 +++ arch/sandbox/dts/test.dts | 23 + configs/sandbox_defconfig | 3 + drivers/clk/Kconfig | 7 + drivers/clk/Makefile | 1 + drivers/clk/clk_rpmi.c | 346 +++++++++++++++ drivers/firmware/Kconfig | 1 + drivers/firmware/Makefile | 1 + drivers/firmware/rpmi/Kconfig | 27 ++ drivers/firmware/rpmi/Makefile | 4 + drivers/firmware/rpmi/rpmi-sandbox-clock.c | 446 +++++++++++++++++++ drivers/firmware/rpmi/rpmi-sandbox-power.c | 174 ++++++++ drivers/firmware/rpmi/rpmi-sandbox.c | 158 +++++++ drivers/firmware/rpmi/rpmi-sandbox.h | 48 +++ drivers/firmware/rpmi/rpmi-sbi-mpxy.c | 449 +++++++++++++++++++ drivers/firmware/rpmi/rpmi-shmem.c | 474 +++++++++++++++++++++ drivers/firmware/rpmi/rpmi-uclass.c | 138 ++++++ drivers/power/domain/Kconfig | 7 + drivers/power/domain/Makefile | 1 + drivers/power/domain/rpmi-power-domain.c | 229 ++++++++++ include/dm/uclass-id.h | 1 + include/rpmi-uclass.h | 79 ++++ include/rpmi.h | 159 +++++++ include/rpmi_proto.h | 143 +++++++ test/dm/Makefile | 1 + test/dm/rpmi.c | 69 +++ 27 files changed, 3055 insertions(+) create mode 100644 drivers/clk/clk_rpmi.c create mode 100644 drivers/firmware/rpmi/Kconfig create mode 100644 drivers/firmware/rpmi/Makefile create mode 100644 drivers/firmware/rpmi/rpmi-sandbox-clock.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox-power.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox.h create mode 100644 drivers/firmware/rpmi/rpmi-sbi-mpxy.c create mode 100644 drivers/firmware/rpmi/rpmi-shmem.c create mode 100644 drivers/firmware/rpmi/rpmi-uclass.c create mode 100644 drivers/power/domain/rpmi-power-domain.c create mode 100644 include/rpmi-uclass.h create mode 100644 include/rpmi.h create mode 100644 include/rpmi_proto.h create mode 100644 test/dm/rpmi.c
Quoted 3 lines
--
2.47.3
Regards,
Anup
Quoted 113 lines · Anup Patel
Quoted 100 lines · Charles Perry
Hello,
This series adds support for RISC-V Platform Management Interface (RPMI) to U-Boot. RPMI is an OS-agnostic protocol for communication between an Application Processor (AP) and a Platform Microcontroller (PuC) [1]. The goals and purpose of RPMI are similar to ARM's SCMI.
This patchset first adds an abstraction layer for the transport (
UCLASS_RPMI), then adds two transport layer (SBI MPXY and shared memory) and finally adds two client, or service group in RPMI jargon: Clock and device power service groups.This series also include a sandbox mode for which the RPMI transport doesn't attach to a firmware but contains a simulation of a firmware. This reuses code from librpmi [2] to create the message responses.
Apart from the spec [1], this series reuses code from the Linux SBI MPXY driver [3], OpenSBI shared memory transport [4] and Linux RPMI clock driver [5].
This was tested on a PIC64-HPSC SoC which features an application cluster and a system controller with isolated address spaces, hence the need for RPMI. I've tested the interoperability between RPMI clients and transports with an S-Mode U-boot (using the MPXY transport) and M-Mode U-Boot (using shared memory transport) and the following device tree fragment:
``` firmware { rpmi: mailbox@7000000000 { compatible = "riscv,rpmi-shmem-mbox"; reg = <0x70 0x00000000 0x0 0x3000>, <0x70 0x00003000 0x0 0x3000>, <0x70 0x00006000 0x0 0x1000>, <0x70 0x00007000 0x0 0x1000>; reg-names = "a2p-req", "p2a-ack", "p2a-req", "a2p-ack"; riscv,slot-size = <64>; #mbox-cells = <1>; }; ifndef CONFIG_RISCV_MMODE clock-service { compatible = "riscv,rpmi-mpxy-clock"; mboxes = <&rpmi 0x8>; riscv,sbi-mpxy-channel-id = <0x1000>; }; device-power-service { compatible = "riscv,rpmi-mpxy-device-power"; mboxes = <&rpmi 0x9>; riscv,sbi-mpxy-channel-id = <0x1001>; }; mpxy_mbox: mailbox { compatible = "riscv,sbi-mpxy-mbox"; #mbox-cells = <2>; }; endif sysc_clk: clock-controller { compatible = "riscv,rpmi-clock"; ifdef CONFIG_RISCV_MMODE mboxes = <&rpmi 0x8>; else mboxes = <&mpxy_mbox 0x1000 0x0>; endif #clock-cells = <1>; }; sysc_dpwr: power-domain-controller { compatible = "riscv,rpmi-device-power"; ifdef CONFIG_RISCV_MMODE mboxes = <&rpmi 0x9>; else mboxes = <&mpxy_mbox 0x1001 0x0>; endif #power-domain-cells = <1>; }; }; ```[1]: https://github.com/riscv-non-isa/riscv-rpmi
[2]: https://github.com/riscv-software-src/librpmi
[3]: https://elixir.bootlin.com/linux/v7.0.10/source/drivers/mailbox/riscv-sbi-mpxy-mbox.c
[4]: https://elixir.bootlin.com/opensbi/v1.8.1/source/lib/utils/mailbox/fdt_mailbox_rpmi_shmem.c
[5]: https://elixir.bootlin.com/linux/v7.0.10/source/drivers/clk/clk-rpmi.cCc Rahul Pathak <[email protected]>Cc Anup Patel <[email protected]>Cc Paul Walmsley <[email protected]>Cc Palmer Dabbelt <[email protected]>Cc Albert Ou <[email protected]>Cc Alexandre Ghiti <[email protected]>Cc Romain Caritey <[email protected]>Cc Mame Maria Mbaye <[email protected]>Cc Tom Rini <[email protected]>Cc Heinrich Schuchardt <[email protected]>Cc Rick Chen <[email protected]>Cc Leo Yu-Chi Liang <[email protected]>Quoted 8 lines
Charles Perry (7):
firmware: add support for RPMI
firmware: rpmi: add support for the SBI MPXY transport
firmware: rpmi: add support for shared memory transport
drivers: clk: add support for RPMI clocks
drivers: power: add support for RPMI power domains
firmware: rpmi: add a test and a sandbox driver
firmware: rpmi: add a test and sandbox for device powerI have a more general question.
Quoted 7 lines
Why add separate firmware framework for RPMI ?
You can just extend the existing U-Boot mailbox framework and just like we have in Linux. The separate transport drivers will be mailbox-controller drivers and each service group will have a separate mailbox-client driver under appropriate driver subsystem.
The mailbox framework of U-Boot is much simpler than the one of Linux and there's no real incentive to have RPMI piggyback on it. The extra code needed to do function dispatch with an enum and a union (struct rpmi_mbox_message [1] and this function [2] in Linux) so that everything can be funneled into the ->send_data() callback would add more churn than creating a new RPMI framework separate from mailbox. rpmi-uclass.c is 138 lines of code which is what I call the RPMI framework, which is about as much code as the definitions and code for the function dispatch code I mentionned above.
[1]: https://elixir.bootlin.com/linux/v7.1.2/source/include/linux/mailbox/riscv-rpmi-message.h#L144
[2]: https://elixir.bootlin.com/linux/v7.1.2/source/drivers/mailbox/riscv-sbi-mpxy-mbox.c#L400
Thanks,
Charles
Quoted 52 lines
Quoted 46 lines
MAINTAINERS | 9 + arch/riscv/include/asm/sbi.h | 57 +++ arch/sandbox/dts/test.dts | 23 + configs/sandbox_defconfig | 3 + drivers/clk/Kconfig | 7 + drivers/clk/Makefile | 1 + drivers/clk/clk_rpmi.c | 346 +++++++++++++++ drivers/firmware/Kconfig | 1 + drivers/firmware/Makefile | 1 + drivers/firmware/rpmi/Kconfig | 27 ++ drivers/firmware/rpmi/Makefile | 4 + drivers/firmware/rpmi/rpmi-sandbox-clock.c | 446 +++++++++++++++++++ drivers/firmware/rpmi/rpmi-sandbox-power.c | 174 ++++++++ drivers/firmware/rpmi/rpmi-sandbox.c | 158 +++++++ drivers/firmware/rpmi/rpmi-sandbox.h | 48 +++ drivers/firmware/rpmi/rpmi-sbi-mpxy.c | 449 +++++++++++++++++++ drivers/firmware/rpmi/rpmi-shmem.c | 474 +++++++++++++++++++++ drivers/firmware/rpmi/rpmi-uclass.c | 138 ++++++ drivers/power/domain/Kconfig | 7 + drivers/power/domain/Makefile | 1 + drivers/power/domain/rpmi-power-domain.c | 229 ++++++++++ include/dm/uclass-id.h | 1 + include/rpmi-uclass.h | 79 ++++ include/rpmi.h | 159 +++++++ include/rpmi_proto.h | 143 +++++++ test/dm/Makefile | 1 + test/dm/rpmi.c | 69 +++ 27 files changed, 3055 insertions(+) create mode 100644 drivers/clk/clk_rpmi.c create mode 100644 drivers/firmware/rpmi/Kconfig create mode 100644 drivers/firmware/rpmi/Makefile create mode 100644 drivers/firmware/rpmi/rpmi-sandbox-clock.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox-power.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox.c create mode 100644 drivers/firmware/rpmi/rpmi-sandbox.h create mode 100644 drivers/firmware/rpmi/rpmi-sbi-mpxy.c create mode 100644 drivers/firmware/rpmi/rpmi-shmem.c create mode 100644 drivers/firmware/rpmi/rpmi-uclass.c create mode 100644 drivers/power/domain/rpmi-power-domain.c create mode 100644 include/rpmi-uclass.h create mode 100644 include/rpmi.h create mode 100644 include/rpmi_proto.h create mode 100644 test/dm/rpmi.cQuoted 3 lines
--
2.47.3Regards,
Anup