diff mbox series

[v4,2/4] SFH: PCI driver to add support of AMD sensor fusion Hub using HID framework

Message ID 1582779537-25662-3-git-send-email-Sandeep.Singh@amd.com (mailing list archive)
State New, archived
Delegated to: Jiri Kosina
Headers show
Series SFH: Add Support for AMD Sensor Fusion Hub | expand

Commit Message

Sandeep Singh Feb. 27, 2020, 4:58 a.m. UTC
From: Sandeep Singh <sandeep.singh@amd.com>

AMD SFH uses HID over PCIe bus.SFH fw is part of MP2
processor (MP2 which is an ARM® Cortex-M4 core based
co-processor to x86) and it runs on MP2 where in driver resides
on X86. This part of module will communicate with MP2 FW and
provide that data into DRAM

Signed-off-by: Sandeep Singh <sandeep.singh@amd.com>
Signed-off-by: Nehal Shah <Nehal-bakulchandra.Shah@amd.com>
Reported-by: kbuild test robot <lkp@intel.com>
---
 drivers/hid/Kconfig                    |   2 +
 drivers/hid/Makefile                   |   1 +
 drivers/hid/amd-sfh-hid/Kconfig        |  20 +++
 drivers/hid/amd-sfh-hid/Makefile       |  16 +++
 drivers/hid/amd-sfh-hid/amd_mp2_pcie.c | 243 +++++++++++++++++++++++++++++++++
 drivers/hid/amd-sfh-hid/amd_mp2_pcie.h | 176 ++++++++++++++++++++++++
 6 files changed, 458 insertions(+)
 create mode 100644 drivers/hid/amd-sfh-hid/Kconfig
 create mode 100644 drivers/hid/amd-sfh-hid/Makefile
 create mode 100644 drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
 create mode 100644 drivers/hid/amd-sfh-hid/amd_mp2_pcie.h

Comments

Singh, Sandeep March 9, 2020, 5:13 a.m. UTC | #1
[snipped]

On 2/27/2020 10:28 AM, Sandeep Singh wrote:
> From: Sandeep Singh <sandeep.singh@amd.com>
>
> AMD SFH uses HID over PCIe bus.SFH fw is part of MP2
> processor (MP2 which is an ARM® Cortex-M4 core based
> co-processor to x86) and it runs on MP2 where in driver resides
> on X86. This part of module will communicate with MP2 FW and
> provide that data into DRAM
>
> Signed-off-by: Sandeep Singh <sandeep.singh@amd.com>
> Signed-off-by: Nehal Shah <Nehal-bakulchandra.Shah@amd.com>
> Reported-by: kbuild test robot <lkp@intel.com>

Guys ,

Do you have any updates on this patch series? Can you please make some 
time to review ?
Andy Shevchenko March 27, 2020, 2:55 p.m. UTC | #2
On Thu, Feb 27, 2020 at 7:01 AM Sandeep Singh <Sandeep.Singh@amd.com> wrote:
>
> From: Sandeep Singh <sandeep.singh@amd.com>
>
> AMD SFH uses HID over PCIe bus.SFH fw is part of MP2
> processor (MP2 which is an ARM® Cortex-M4 core based
> co-processor to x86) and it runs on MP2 where in driver resides
> on X86. This part of module will communicate with MP2 FW and
> provide that data into DRAM

You asked for review, here you are.
TL,DR; it requires a bit of work.

...

> +       depends on (X86_64 || COMPILE_TEST) && PCI

For better maintenance
       depends on X86_64 || COMPILE_TEST
       depends on PCI

...

> +# SPDX-License-Identifier: GPL-2.0
> +#
> +# Makefile - AMD SFH HID drivers
> +# Copyright (c) 2020-2021, Advanced Micro Devices, Inc.
> +#

> +#

Perhaps simple blank line instead?

> +ccflags-y += -I$(srctree)/$(src)/

Why?

...

> +#include <linux/interrupt.h>
> +#include <linux/module.h>
> +#include <linux/pci.h>
> +#include <linux/slab.h>
> +#include <linux/delay.h>

Keep in order?

+ blank line?

Missed bits.h, types.h.

> +#include "amd_mp2_pcie.h"

...

> +       write64((u64)info.phy_address, privdata->mmio + AMD_C2P_MSG2);

Why explicit cast?

...

> +       /*fill up command register */

Space is missed.

...

> +       if (!sensor_id)
> +               return -ENOMEM;

I can say -EINVAL as per its definition, but why do you need this at all?

...

> +static int amd_mp2_pci_init(struct amd_mp2_dev *privdata, struct pci_dev *pdev)
> +{
> +       int rc;
> +       int bar_index = 2;
> +       resource_size_t size, base;

> +       pci_set_drvdata(pdev, privdata);

Better to assign when you are sure (to some extend in both of them):
a) it's needed
b) driver is going to be probed correctly

...

> +       rc = pcim_iomap_regions(pdev, 1 >> 2, DRIVER_NAME);

What 1 >> 2 means? Shouldn't be simple BIT(2)?
How was this been tested?

> +       if (rc)
> +               goto err_pci_regions;

...

> +       base = pci_resource_start(pdev, bar_index);
> +       size = pci_resource_len(pdev, bar_index);
> +       dev_dbg(ndev_dev(privdata), "Base addr:%llx size:%llx\n",
> +               (unsigned long long)base, (unsigned long long)size);

Read printk-formats.rst.
Now, when you get familiar with it, find proper specifier and drop
these ugly castings.
But wait, why do you need this? `dmesg` will show it anyway during
boot / hotplug event time.

...

> +       privdata->mmio = ioremap(base, size);
> +       if (!privdata->mmio) {
> +               rc = -EIO;
> +               goto err_dma_mask;
> +       }

Why?!

...

> +err_dma_mask:
> +       pci_clear_master(pdev);
> +err_pci_regions:
> +       pci_disable_device(pdev);

Are you using devres or not? Please, be consistent.

> +err_pci_enable:

> +       pci_set_drvdata(pdev, NULL);

I think it's some like 5 to 10 years that we don't need this.

> +       return rc;
> +}

...

> +       pci_iounmap(pdev, privdata->mmio);
> +
> +       pci_clear_master(pdev);
> +       pci_disable_device(pdev);
> +       pci_set_drvdata(pdev, NULL);

Ditto as above two comments.

...

> +       dev_info(&pdev->dev, "MP2 device found [%04x:%04x] (rev %x)\n",
> +                (int)pdev->vendor, (int)pdev->device, (int)pdev->revision);

Oh, if you use explicit casting for printf(), 99.9% you are doing
something wrong (in kernel space).
On top of that, why this noise is here?

...

> +       privdata = devm_kzalloc(&pdev->dev, sizeof(*privdata), GFP_KERNEL);

> +

No need for this blank line.

> +       if (!privdata) {

> +       }

...


> +       rc = amd_mp2_pci_init(privdata, pdev);
> +       if (rc)
> +               goto err_pci_init;
> +       return 0;

Why its content can't be simple here? I.o.w. why this function is needed?

...

> +err_pci_init:
> +       return rc;
> +err_dev:
> +       return rc;

Completely useless code.

> +}

...

> +static const struct pci_device_id amd_mp2_pci_tbl[] = {
> +       {PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_MP2)},
> +       {0}

0 is not needed, but it's up to you.

> +};

...

> +static int __init amd_mp2_pci_driver_init(void)
> +{
> +       return pci_register_driver(&amd_mp2_pci_driver);
> +}
> +module_init(amd_mp2_pci_driver_init);
> +
> +static void __exit amd_mp2_pci_driver_exit(void)
> +{
> +       pci_unregister_driver(&amd_mp2_pci_driver);
> +}
> +module_exit(amd_mp2_pci_driver_exit);

module_pci_driver()
We have it for years.

...

> +#include <linux/pci.h>

I don't see users of it, but missed headers
types.h

...

> +#define PCI_DEVICE_ID_AMD_MP2  0x15E4

Why it's not in C file?

...

> +#define AMD_P2C_MSG0   0x10680 /*Do not use*/
> +#define AMD_P2C_MSG1   0x10684
> +#define AMD_P2C_MSG2   0x10688
> +#define AMD_P2C_MSG3   0x1068C /*MP2 debug info*/
> +#define AMD_P2C_MSG_INTEN      0x10690 /*MP2 int gen register*/
> +#define AMD_P2C_MSG_INTSTS     0x10694 /*Interrupt sts*/

Missed spaces.

...

> +#define write64 amdsfh_write64
> +static inline void amdsfh_write64(u64 val, void __iomem *mmio)
> +{
> +       writel(val, mmio);
> +       writel(val >> 32, mmio + sizeof(u32));
> +}

NIH of lo_hi_writeq().

> +#define read64 amdsfh_read64
> +static inline u64 amdsfh_read64(void __iomem *mmio)
> +{
> +       u64 low, high;
> +
> +       low = readl(mmio);
> +       high = readl(mmio + sizeof(u32));
> +       return low | (high << 32);
> +}

NIH of lo_hi_readq().

...

> +struct sfh_command_register {
> +       union sfh_cmd_base cmd_base;
> +       union sfh_command_parameter cmd_param;

> +       phys_addr_t phy_addr;

Are you sure? This type is flexible. And by name of the struct I think
it operates with hardware, so, fix it accordingly.

> +};

...

> +enum response_type {
> +       non_operationevent,
> +       command_success,
> +       command_failed,
> +       sfi_dataready_event,
> +       invalid_response = 0xff,

GENMASK()

> +};

UPPER CASE?

> +enum status_type {
> +       cmd_success,
> +       invalid_data_payload,
> +       invalid_data_length,
> +       invalid_sensor_id,
> +       invalid_dram_addr,
> +       invalid_command,
> +       sensor_enabled,
> +       sensor_disabled,
> +       status_end,
> +};
> +
> +enum command_id {
> +       non_operation = 0,
> +       enable_sensor = 1,
> +       disable_sensor = 2,
> +       dump_sensorinfo = 3,
> +       numberof_sensordiscovered = 4,
> +       who_am_i_regchipid = 5,
> +       set_dcd_data = 6,
> +       get_dcd_data = 7,
> +       stop_all_sensors = 8,
> +       invalid_cmd = 0xf,
> +};

Ditto.

...

> +enum sensor_idx {

Do you need names for enums like this?

> +       ACCEL_IDX               = 0,
> +       GYRO_IDX                = 1,
> +       MAG_IDX                 = 2,
> +       AMBIENT_LIGHT_IDX       = 19,
> +       NUM_ALL_SENSOR_CONSUMERS
> +};

...

> +struct amd_mp2_dev {
> +       struct pci_dev *pdev;

> +       void __iomem *mmio;

Header for __iomem?

> +       struct delayed_work work;

Header for this?

> +       void *ctx;
> +       void *cl_data;
> +};

...

> +struct amd_mp2_sensor_info {
> +       u8 sensor_idx;
> +       u32 period;

> +       phys_addr_t phy_address;

Same comment as per above use of phys_addr_t type.

> +};

...

> +#define ndev_pdev(ndev) ((ndev)->pdev)
> +#define ndev_name(ndev) pci_name(ndev_pdev(ndev))
> +#define ndev_dev(ndev) (&ndev_pdev(ndev)->dev)

Why? What's the benefit?
Singh, Sandeep March 30, 2020, 5:12 a.m. UTC | #3
Thanks Andy for review comments i will look into it.

On 3/27/2020 8:25 PM, Andy Shevchenko wrote:
> [CAUTION: External Email]
>
> On Thu, Feb 27, 2020 at 7:01 AM Sandeep Singh <Sandeep.Singh@amd.com> wrote:
>> From: Sandeep Singh <sandeep.singh@amd.com>
>>
>> AMD SFH uses HID over PCIe bus.SFH fw is part of MP2
>> processor (MP2 which is an ARM® Cortex-M4 core based
>> co-processor to x86) and it runs on MP2 where in driver resides
>> on X86. This part of module will communicate with MP2 FW and
>> provide that data into DRAM
> You asked for review, here you are.
> TL,DR; it requires a bit of work.
>
> ...
>
>> +       depends on (X86_64 || COMPILE_TEST) && PCI
> For better maintenance
>         depends on X86_64 || COMPILE_TEST
>         depends on PCI
>
> ...
>
>> +# SPDX-License-Identifier: GPL-2.0
>> +#
>> +# Makefile - AMD SFH HID drivers
>> +# Copyright (c) 2020-2021, Advanced Micro Devices, Inc.
>> +#
>> +#
> Perhaps simple blank line instead?
>
>> +ccflags-y += -I$(srctree)/$(src)/
> Why?
>
> ...
>
>> +#include <linux/interrupt.h>
>> +#include <linux/module.h>
>> +#include <linux/pci.h>
>> +#include <linux/slab.h>
>> +#include <linux/delay.h>
> Keep in order?
>
> + blank line?
>
> Missed bits.h, types.h.
>
>> +#include "amd_mp2_pcie.h"
> ...
>
>> +       write64((u64)info.phy_address, privdata->mmio + AMD_C2P_MSG2);
> Why explicit cast?
>
> ...
>
>> +       /*fill up command register */
> Space is missed.
>
> ...
>
>> +       if (!sensor_id)
>> +               return -ENOMEM;
> I can say -EINVAL as per its definition, but why do you need this at all?
>
> ...
>
>> +static int amd_mp2_pci_init(struct amd_mp2_dev *privdata, struct pci_dev *pdev)
>> +{
>> +       int rc;
>> +       int bar_index = 2;
>> +       resource_size_t size, base;
>> +       pci_set_drvdata(pdev, privdata);
> Better to assign when you are sure (to some extend in both of them):
> a) it's needed
> b) driver is going to be probed correctly
>
> ...
>
>> +       rc = pcim_iomap_regions(pdev, 1 >> 2, DRIVER_NAME);
> What 1 >> 2 means? Shouldn't be simple BIT(2)?
> How was this been tested?
>
>> +       if (rc)
>> +               goto err_pci_regions;
> ...
>
>> +       base = pci_resource_start(pdev, bar_index);
>> +       size = pci_resource_len(pdev, bar_index);
>> +       dev_dbg(ndev_dev(privdata), "Base addr:%llx size:%llx\n",
>> +               (unsigned long long)base, (unsigned long long)size);
> Read printk-formats.rst.
> Now, when you get familiar with it, find proper specifier and drop
> these ugly castings.
> But wait, why do you need this? `dmesg` will show it anyway during
> boot / hotplug event time.
>
> ...
>
>> +       privdata->mmio = ioremap(base, size);
>> +       if (!privdata->mmio) {
>> +               rc = -EIO;
>> +               goto err_dma_mask;
>> +       }
> Why?!
>
> ...
>
>> +err_dma_mask:
>> +       pci_clear_master(pdev);
>> +err_pci_regions:
>> +       pci_disable_device(pdev);
> Are you using devres or not? Please, be consistent.
>
>> +err_pci_enable:
>> +       pci_set_drvdata(pdev, NULL);
> I think it's some like 5 to 10 years that we don't need this.
>
>> +       return rc;
>> +}
> ...
>
>> +       pci_iounmap(pdev, privdata->mmio);
>> +
>> +       pci_clear_master(pdev);
>> +       pci_disable_device(pdev);
>> +       pci_set_drvdata(pdev, NULL);
> Ditto as above two comments.
>
> ...
>
>> +       dev_info(&pdev->dev, "MP2 device found [%04x:%04x] (rev %x)\n",
>> +                (int)pdev->vendor, (int)pdev->device, (int)pdev->revision);
> Oh, if you use explicit casting for printf(), 99.9% you are doing
> something wrong (in kernel space).
> On top of that, why this noise is here?
>
> ...
>
>> +       privdata = devm_kzalloc(&pdev->dev, sizeof(*privdata), GFP_KERNEL);
>> +
> No need for this blank line.
>
>> +       if (!privdata) {
>> +       }
> ...
>
>
>> +       rc = amd_mp2_pci_init(privdata, pdev);
>> +       if (rc)
>> +               goto err_pci_init;
>> +       return 0;
> Why its content can't be simple here? I.o.w. why this function is needed?
>
> ...
>
>> +err_pci_init:
>> +       return rc;
>> +err_dev:
>> +       return rc;
> Completely useless code.
>
>> +}
> ...
>
>> +static const struct pci_device_id amd_mp2_pci_tbl[] = {
>> +       {PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_MP2)},
>> +       {0}
> 0 is not needed, but it's up to you.
>
>> +};
> ...
>
>> +static int __init amd_mp2_pci_driver_init(void)
>> +{
>> +       return pci_register_driver(&amd_mp2_pci_driver);
>> +}
>> +module_init(amd_mp2_pci_driver_init);
>> +
>> +static void __exit amd_mp2_pci_driver_exit(void)
>> +{
>> +       pci_unregister_driver(&amd_mp2_pci_driver);
>> +}
>> +module_exit(amd_mp2_pci_driver_exit);
> module_pci_driver()
> We have it for years.
>
> ...
>
>> +#include <linux/pci.h>
> I don't see users of it, but missed headers
> types.h
>
> ...
>
>> +#define PCI_DEVICE_ID_AMD_MP2  0x15E4
> Why it's not in C file?
>
> ...
>
>> +#define AMD_P2C_MSG0   0x10680 /*Do not use*/
>> +#define AMD_P2C_MSG1   0x10684
>> +#define AMD_P2C_MSG2   0x10688
>> +#define AMD_P2C_MSG3   0x1068C /*MP2 debug info*/
>> +#define AMD_P2C_MSG_INTEN      0x10690 /*MP2 int gen register*/
>> +#define AMD_P2C_MSG_INTSTS     0x10694 /*Interrupt sts*/
> Missed spaces.
>
> ...
>
>> +#define write64 amdsfh_write64
>> +static inline void amdsfh_write64(u64 val, void __iomem *mmio)
>> +{
>> +       writel(val, mmio);
>> +       writel(val >> 32, mmio + sizeof(u32));
>> +}
> NIH of lo_hi_writeq().
>
>> +#define read64 amdsfh_read64
>> +static inline u64 amdsfh_read64(void __iomem *mmio)
>> +{
>> +       u64 low, high;
>> +
>> +       low = readl(mmio);
>> +       high = readl(mmio + sizeof(u32));
>> +       return low | (high << 32);
>> +}
> NIH of lo_hi_readq().
>
> ...
>
>> +struct sfh_command_register {
>> +       union sfh_cmd_base cmd_base;
>> +       union sfh_command_parameter cmd_param;
>> +       phys_addr_t phy_addr;
> Are you sure? This type is flexible. And by name of the struct I think
> it operates with hardware, so, fix it accordingly.
>
>> +};
> ...
>
>> +enum response_type {
>> +       non_operationevent,
>> +       command_success,
>> +       command_failed,
>> +       sfi_dataready_event,
>> +       invalid_response = 0xff,
> GENMASK()
>
>> +};
> UPPER CASE?
>
>> +enum status_type {
>> +       cmd_success,
>> +       invalid_data_payload,
>> +       invalid_data_length,
>> +       invalid_sensor_id,
>> +       invalid_dram_addr,
>> +       invalid_command,
>> +       sensor_enabled,
>> +       sensor_disabled,
>> +       status_end,
>> +};
>> +
>> +enum command_id {
>> +       non_operation = 0,
>> +       enable_sensor = 1,
>> +       disable_sensor = 2,
>> +       dump_sensorinfo = 3,
>> +       numberof_sensordiscovered = 4,
>> +       who_am_i_regchipid = 5,
>> +       set_dcd_data = 6,
>> +       get_dcd_data = 7,
>> +       stop_all_sensors = 8,
>> +       invalid_cmd = 0xf,
>> +};
> Ditto.
>
> ...
>
>> +enum sensor_idx {
> Do you need names for enums like this?
>
>> +       ACCEL_IDX               = 0,
>> +       GYRO_IDX                = 1,
>> +       MAG_IDX                 = 2,
>> +       AMBIENT_LIGHT_IDX       = 19,
>> +       NUM_ALL_SENSOR_CONSUMERS
>> +};
> ...
>
>> +struct amd_mp2_dev {
>> +       struct pci_dev *pdev;
>> +       void __iomem *mmio;
> Header for __iomem?
>
>> +       struct delayed_work work;
> Header for this?
>
>> +       void *ctx;
>> +       void *cl_data;
>> +};
> ...
>
>> +struct amd_mp2_sensor_info {
>> +       u8 sensor_idx;
>> +       u32 period;
>> +       phys_addr_t phy_address;
> Same comment as per above use of phys_addr_t type.
>
>> +};
> ...
>
>> +#define ndev_pdev(ndev) ((ndev)->pdev)
>> +#define ndev_name(ndev) pci_name(ndev_pdev(ndev))
>> +#define ndev_dev(ndev) (&ndev_pdev(ndev)->dev)
> Why? What's the benefit?
>
> --
> With Best Regards,
> Andy Shevchenko

Regards

Sandeep
Richard Neumann March 30, 2020, 8:33 a.m. UTC | #4
I took some time testing the patch series on my HP ENVY x360 13-
ag0005ng with a Ryzen 5 2500U.
On my machine the readout of register AMD_P2C_MSG3 (0x1068C) always
comes back as zero (0x0) and thusly activestatus will be zero as well.
Consequently the bistmask (0x0) will not match any sensor types.
I came up with a patch [1] on top of this patch series to add module
parameters to be able to statically enable the different sensors.
If I load the module with "force_accel=1" I get a working accelerometer
HID device and working screen rotation through iio-sensor-proxy. This
only works so far, if I boot the kernel with "amd_iommu=off", which
seems to be a general issue with this driver [2],[3].

[1] https://gist.githubusercontent.com/conqp/e8a0793406fbe7c9714f01f3078ea33a/raw/ac2de743e68d1dd90430e57cb28df8c1dc5d1098/amd-sfh.patch
[2] https://bbs.archlinux.org/viewtopic.php?id=253058
[3] https://bbs.archlinux.org/viewtopic.php?id=252815
Richard Neumann March 31, 2020, 12:31 p.m. UTC | #5
Not a real review, but your patch series seems to be repeating a lot
from drivers/i2c/busses/i2c-amd-mp2*.
Is there any chance we could re-use the code?
E.g. the AMD_C2P_* definitions from drivers/i2c/busses/i2c-amd-mp2.h?

Am Donnerstag, den 27.02.2020, 10:28 +0530 schrieb Sandeep Singh:
> From: Sandeep Singh <sandeep.singh@amd.com>
> 
> AMD SFH uses HID over PCIe bus.SFH fw is part of MP2
> processor (MP2 which is an ARM® Cortex-M4 core based
> co-processor to x86) and it runs on MP2 where in driver resides
> on X86. This part of module will communicate with MP2 FW and
> provide that data into DRAM
> 
> Signed-off-by: Sandeep Singh <sandeep.singh@amd.com>
> Signed-off-by: Nehal Shah <Nehal-bakulchandra.Shah@amd.com>
> Reported-by: kbuild test robot <lkp@intel.com>
> ---
>  drivers/hid/Kconfig                    |   2 +
>  drivers/hid/Makefile                   |   1 +
>  drivers/hid/amd-sfh-hid/Kconfig        |  20 +++
>  drivers/hid/amd-sfh-hid/Makefile       |  16 +++
>  drivers/hid/amd-sfh-hid/amd_mp2_pcie.c | 243
> +++++++++++++++++++++++++++++++++
>  drivers/hid/amd-sfh-hid/amd_mp2_pcie.h | 176
> ++++++++++++++++++++++++
>  6 files changed, 458 insertions(+)
>  create mode 100644 drivers/hid/amd-sfh-hid/Kconfig
>  create mode 100644 drivers/hid/amd-sfh-hid/Makefile
>  create mode 100644 drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
>  create mode 100644 drivers/hid/amd-sfh-hid/amd_mp2_pcie.h
> 
> diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig
> index 494a39e..b253ad1 100644
> --- a/drivers/hid/Kconfig
> +++ b/drivers/hid/Kconfig
> @@ -1155,4 +1155,6 @@ source "drivers/hid/i2c-hid/Kconfig"
>  
>  source "drivers/hid/intel-ish-hid/Kconfig"
>  
> +source "drivers/hid/amd-sfh-hid/Kconfig"
> +
>  endmenu
> diff --git a/drivers/hid/Makefile b/drivers/hid/Makefile
> index bfefa36..15a08e8 100644
> --- a/drivers/hid/Makefile
> +++ b/drivers/hid/Makefile
> @@ -139,3 +139,4 @@ obj-$(CONFIG_I2C_HID)		+= i2c-hid/
>  
>  obj-$(CONFIG_INTEL_ISH_HID)	+= intel-ish-hid/
>  obj-$(INTEL_ISH_FIRMWARE_DOWNLOADER)	+= intel-ish-hid/
> +obj-$(CONFIG_AMD_SFH_HID)       += amd-sfh-hid/
> diff --git a/drivers/hid/amd-sfh-hid/Kconfig b/drivers/hid/amd-sfh-
> hid/Kconfig
> new file mode 100644
> index 0000000..7a224a1
> --- /dev/null
> +++ b/drivers/hid/amd-sfh-hid/Kconfig
> @@ -0,0 +1,20 @@
> +# SPDX-License-Identifier: GPL-2.0-only
> +menu "AMD SFH HID support"
> +	depends on (X86_64 || COMPILE_TEST) && PCI
> +
> +config AMD_SFH_HID
> +	tristate "AMD Sensor Fusion Hub"
> +	select HID
> +	help
> +	If you say yes to this option, support will be included for the
> AMD
> +	Sensor Fusion Hub.
> +	This driver will enable sensors functionality to user through
> HID
> +	framework. Basically this driver will get data from MP2 FW
> +	and provide that data to HID framework.
> +	MP2 which is an ARM® Cortex-M4 core based co-processor to x86.
> +
> +	This driver can also be built as modules. If so, the modules
> will
> +	be  called amd-mp2-pcie and amd-sfhtp-hid.
> +	Say Y or M here if you want to support AMD SFH. If unsure, say
> N.
> +
> +endmenu
> diff --git a/drivers/hid/amd-sfh-hid/Makefile b/drivers/hid/amd-sfh-
> hid/Makefile
> new file mode 100644
> index 0000000..fa38d84
> --- /dev/null
> +++ b/drivers/hid/amd-sfh-hid/Makefile
> @@ -0,0 +1,16 @@
> +# SPDX-License-Identifier: GPL-2.0
> +#
> +# Makefile - AMD SFH HID drivers
> +# Copyright (c) 2020-2021, Advanced Micro Devices, Inc.
> +#
> +#
> +ccflags-m := -Werror
> +obj-$(CONFIG_AMD_SFH_HID) += amd-mp2-pcie.o
> +amd-mp2-pcie-objs := amd_mp2_pcie.o
> +
> +obj-$(CONFIG_AMD_SFH_HID) +=amd-sfhtp-hid.o
> +amd-sfhtp-hid-objs := amdsfh-hid.o
> +amd-sfhtp-hid-objs+= amdsfh-hid-client.o
> +amd-sfhtp-hid-objs+= hid_descriptor/amd_sfh_hid_descriptor.o
> +
> +ccflags-y += -I$(srctree)/$(src)/
> diff --git a/drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
> b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
> new file mode 100644
> index 0000000..c67f389
> --- /dev/null
> +++ b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
> @@ -0,0 +1,243 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * AMD MP2 PCIe communication driver
> + *
> + * Authors: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
> + *          Nehal Bakulchandra Shah <Nehal-bakulchandra.Shah@amd.com
> >
> + */
> +
> +#include <linux/interrupt.h>
> +#include <linux/module.h>
> +#include <linux/pci.h>
> +#include <linux/slab.h>
> +#include <linux/delay.h>
> +#include "amd_mp2_pcie.h"
> +
> +#define DRIVER_NAME	"pcie_mp2_amd"
> +#define DRIVER_DESC	"AMD(R) PCIe MP2 Communication Driver"
> +
> +#define	ACEL_EN		BIT(ACCEL_IDX)
> +#define	GYRO_EN		BIT(GYRO_IDX)
> +#define MAGNO_EN	BIT(MAG_IDX)
> +#define ALS_EN		BIT(AMBIENT_LIGHT_IDX)
> +
> +void amd_start_sensor(struct pci_dev *pdev, struct
> amd_mp2_sensor_info info)
> +{
> +	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
> +	union sfh_command_parameter cmd_param;
> +	union sfh_cmd_base cmd_base;
> +
> +	/*fill up command register*/
> +	cmd_base.ul = 0;
> +	cmd_base.s.cmd_id = enable_sensor;
> +	cmd_base.s.period = info.period;
> +	cmd_base.s.sensor_id = info.sensor_idx;
> +
> +	/*fill up command param register*/
> +	cmd_param.ul = 0;
> +	cmd_param.s.buffer_layout = 1;
> +	cmd_param.s.buffer_length = 16;
> +
> +	write64((u64)info.phy_address, privdata->mmio + AMD_C2P_MSG2);
> +	writel(cmd_param.ul, privdata->mmio + AMD_C2P_MSG1);
> +	writel(cmd_base.ul, privdata->mmio + AMD_C2P_MSG0);
> +}
> +EXPORT_SYMBOL_GPL(amd_start_sensor);
> +
> +void amd_stop_sensor(struct pci_dev *pdev, u16 sensor_idx)
> +{
> +	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
> +	union sfh_cmd_base cmd_base;
> +
> +	/* fill up command register */
> +	cmd_base.ul = 0;
> +	cmd_base.s.cmd_id = disable_sensor;
> +	cmd_base.s.period = 0;
> +	cmd_base.s.sensor_id = sensor_idx;
> +
> +	write64(0x0, privdata->mmio + AMD_C2P_MSG2);
> +	writel(cmd_base.ul, privdata->mmio + AMD_C2P_MSG0);
> +}
> +EXPORT_SYMBOL_GPL(amd_stop_sensor);
> +
> +void amd_stop_all_sensors(struct pci_dev *pdev)
> +{
> +	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
> +	union sfh_cmd_base cmd_base;
> +
> +	/*fill up command register */
> +	cmd_base.ul = 0;
> +	cmd_base.s.cmd_id = stop_all_sensors;
> +	cmd_base.s.period = 0;
> +	cmd_base.s.sensor_id = 0;
> +
> +	writel(cmd_base.ul, privdata->mmio + AMD_C2P_MSG0);
> +}
> +EXPORT_SYMBOL_GPL(amd_stop_all_sensors);
> +
> +int amd_mp2_get_sensor_num(struct pci_dev *dev, u8 *sensor_id)
> +{
> +	struct amd_mp2_dev *privdata = pci_get_drvdata(dev);
> +	int activestatus;
> +	int num_of_sensors = 0;
> +
> +	if (!sensor_id)
> +		return -ENOMEM;
> +
> +	privdata->eventreg.activecontrolstatus =
> +			readl(privdata->mmio + AMD_P2C_MSG3);
> +	activestatus = privdata->eventreg.activecontrolstatus >> 4;
> +
> +	if (ACEL_EN  & activestatus) {
> +		sensor_id[num_of_sensors] = ACCEL_IDX;
> +		num_of_sensors++;
> +	}
> +	if (GYRO_EN & activestatus) {
> +		sensor_id[num_of_sensors] = GYRO_IDX;
> +		num_of_sensors++;
> +	}
> +	if (MAGNO_EN & activestatus) {
> +		sensor_id[num_of_sensors] = MAG_IDX;
> +		num_of_sensors++;
> +	}
> +
> +	if (ALS_EN & activestatus) {
> +		sensor_id[num_of_sensors] = AMBIENT_LIGHT_IDX;
> +		num_of_sensors++;
> +	}
> +
> +	return num_of_sensors;
> +}
> +EXPORT_SYMBOL_GPL(amd_mp2_get_sensor_num);
> +
> +static int amd_mp2_pci_init(struct amd_mp2_dev *privdata, struct
> pci_dev *pdev)
> +{
> +	int rc;
> +	int bar_index = 2;
> +	resource_size_t size, base;
> +
> +	pci_set_drvdata(pdev, privdata);
> +
> +	rc = pcim_enable_device(pdev);
> +	if (rc)
> +		goto err_pci_enable;
> +
> +	rc = pcim_iomap_regions(pdev, 1 >> 2, DRIVER_NAME);
> +	if (rc)
> +		goto err_pci_regions;
> +
> +	privdata->mmio = pcim_iomap_table(pdev)[2];
> +	pci_set_master(pdev);
> +
> +	rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
> +	if (rc) {
> +		rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
> +		if (rc)
> +			goto err_dma_mask;
> +	}
> +
> +	rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
> +	if (rc) {
> +		rc = pci_set_consistent_dma_mask(pdev,
> DMA_BIT_MASK(32));
> +		if (rc)
> +			goto err_dma_mask;
> +	}
> +
> +	base = pci_resource_start(pdev, bar_index);
> +	size = pci_resource_len(pdev, bar_index);
> +	dev_dbg(ndev_dev(privdata), "Base addr:%llx size:%llx\n",
> +		(unsigned long long)base, (unsigned long long)size);
> +
> +	privdata->mmio = ioremap(base, size);
> +	if (!privdata->mmio) {
> +		rc = -EIO;
> +		goto err_dma_mask;
> +	}
> +
> +	return 0;
> +
> +err_dma_mask:
> +	pci_clear_master(pdev);
> +err_pci_regions:
> +	pci_disable_device(pdev);
> +err_pci_enable:
> +	pci_set_drvdata(pdev, NULL);
> +	return rc;
> +}
> +
> +static void amd_mp2_pci_deinit(struct amd_mp2_dev *privdata)
> +{
> +	struct pci_dev *pdev = ndev_pdev(privdata);
> +
> +	amd_stop_all_sensors(pdev);
> +	pci_iounmap(pdev, privdata->mmio);
> +
> +	pci_clear_master(pdev);
> +	pci_disable_device(pdev);
> +	pci_set_drvdata(pdev, NULL);
> +}
> +
> +static int amd_mp2_pci_probe(struct pci_dev *pdev,
> +			     const struct pci_device_id *id)
> +{
> +	struct amd_mp2_dev *privdata;
> +	int rc;
> +
> +	dev_info(&pdev->dev, "MP2 device found [%04x:%04x] (rev %x)\n",
> +		 (int)pdev->vendor, (int)pdev->device, (int)pdev-
> >revision);
> +
> +	privdata = devm_kzalloc(&pdev->dev, sizeof(*privdata),
> GFP_KERNEL);
> +
> +	if (!privdata) {
> +		rc = -ENOMEM;
> +		goto err_dev;
> +	}
> +
> +	privdata->pdev = pdev;
> +
> +	rc = amd_mp2_pci_init(privdata, pdev);
> +	if (rc)
> +		goto err_pci_init;
> +	return 0;
> +
> +err_pci_init:
> +	return rc;
> +err_dev:
> +	return rc;
> +}
> +
> +static void amd_mp2_pci_remove(struct pci_dev *pdev)
> +{
> +	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
> +
> +	amd_mp2_pci_deinit(privdata);
> +}
> +
> +static const struct pci_device_id amd_mp2_pci_tbl[] = {
> +	{PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_MP2)},
> +	{0}
> +};
> +MODULE_DEVICE_TABLE(pci, amd_mp2_pci_tbl);
> +
> +static struct pci_driver amd_mp2_pci_driver = {
> +	.name		= DRIVER_NAME,
> +	.id_table	= amd_mp2_pci_tbl,
> +	.probe		= amd_mp2_pci_probe,
> +	.remove		= amd_mp2_pci_remove,
> +};
> +
> +static int __init amd_mp2_pci_driver_init(void)
> +{
> +	return pci_register_driver(&amd_mp2_pci_driver);
> +}
> +module_init(amd_mp2_pci_driver_init);
> +
> +static void __exit amd_mp2_pci_driver_exit(void)
> +{
> +	pci_unregister_driver(&amd_mp2_pci_driver);
> +}
> +module_exit(amd_mp2_pci_driver_exit);
> +MODULE_DESCRIPTION(DRIVER_DESC);
> +MODULE_LICENSE("Dual BSD/GPL");
> +MODULE_AUTHOR("Shyam Sundar S K <Shyam-sundar.S-k@amd.com>");
> +MODULE_AUTHOR("Nehal Bakulchandra Shah <
> Nehal-bakulchandra.Shah@amd.com>");
> diff --git a/drivers/hid/amd-sfh-hid/amd_mp2_pcie.h
> b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.h
> new file mode 100644
> index 0000000..3ba69ac
> --- /dev/null
> +++ b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.h
> @@ -0,0 +1,176 @@
> +/* SPDX-License-Identifier: GPL-2.0
> + *
> + * AMD MP2 PCIe communication driver
> + *
> + * Authors: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
> + *          Nehal Bakulchandra Shah <Nehal-bakulchandra.Shah@amd.com
> >
> + */
> +
> +#ifndef PCIE_MP2_AMD_H
> +#define PCIE_MP2_AMD_H
> +
> +#include <linux/pci.h>
> +#define PCI_DEVICE_ID_AMD_MP2	0x15E4
> +
> +/* MP2 C2P Message Registers */
> +#define AMD_C2P_MSG0	0x10500
> +#define AMD_C2P_MSG1	0x10504
> +#define AMD_C2P_MSG2	0x10508
> +#define AMD_C2P_MSG3	0x1050c
> +#define AMD_C2P_MSG4	0x10510
> +#define AMD_C2P_MSG5	0x10514
> +#define AMD_C2P_MSG6	0x10518
> +#define AMD_C2P_MSG7	0x1051c
> +#define AMD_C2P_MSG8	0x10520
> +#define AMD_C2P_MSG9	0x10524
> +
> +/* MP2 P2C Message Registers */
> +#define AMD_P2C_MSG0	0x10680 /*Do not use*/
> +#define AMD_P2C_MSG1	0x10684
> +#define AMD_P2C_MSG2	0x10688
> +#define AMD_P2C_MSG3	0x1068C /*MP2 debug info*/
> +#define AMD_P2C_MSG_INTEN	0x10690 /*MP2 int gen register*/
> +#define AMD_P2C_MSG_INTSTS	0x10694 /*Interrupt sts*/
> +
> +#define write64 amdsfh_write64
> +static inline void amdsfh_write64(u64 val, void __iomem *mmio)
> +{
> +	writel(val, mmio);
> +	writel(val >> 32, mmio + sizeof(u32));
> +}
> +
> +#define read64 amdsfh_read64
> +static inline u64 amdsfh_read64(void __iomem *mmio)
> +{
> +	u64 low, high;
> +
> +	low = readl(mmio);
> +	high = readl(mmio + sizeof(u32));
> +	return low | (high << 32);
> +}
> +
> +/*
> + * SFH Command registers
> + */
> +union sfh_cmd_base {
> +	u32 ul;
> +	struct {
> +		u32 cmd_id : 8;
> +		u32 sensor_id : 8;
> +		u32 period : 16;
> +	} s;
> +};
> +
> +union sfh_command_parameter {
> +	u32 ul;
> +	struct {
> +		u32 buffer_layout : 2;
> +		u32 buffer_length : 6;
> +		u32 rsvd : 24;
> +	} s;
> +};
> +
> +struct sfh_command_register {
> +	union sfh_cmd_base cmd_base;
> +	union sfh_command_parameter cmd_param;
> +	phys_addr_t phy_addr;
> +};
> +
> +/*
> + * SFH Response registers
> + */
> +enum response_type {
> +	non_operationevent,
> +	command_success,
> +	command_failed,
> +	sfi_dataready_event,
> +	invalid_response = 0xff,
> +};
> +
> +enum status_type {
> +	cmd_success,
> +	invalid_data_payload,
> +	invalid_data_length,
> +	invalid_sensor_id,
> +	invalid_dram_addr,
> +	invalid_command,
> +	sensor_enabled,
> +	sensor_disabled,
> +	status_end,
> +};
> +
> +enum command_id {
> +	non_operation = 0,
> +	enable_sensor = 1,
> +	disable_sensor = 2,
> +	dump_sensorinfo = 3,
> +	numberof_sensordiscovered = 4,
> +	who_am_i_regchipid = 5,
> +	set_dcd_data = 6,
> +	get_dcd_data = 7,
> +	stop_all_sensors = 8,
> +	invalid_cmd = 0xf,
> +};
> +
> +/**
> + * union sfh_event_base : bit access of C2P commands
> + * @response: bit: 0..3 SFI response_type
> + * @status: bit: 6..5 status_type
> + * @out_in_c2p: bit: 5 0- output in DRAM,1-in C2PMsg
> + * @length: bit: 8..13 length
> + * @dbg:bit: 14.15 dbg msg include in p2c msg 1-2
> + * @sensor_id:bit: 16..23 Sensor ID
> + * @rsvd:bit: 24..31 Reservered for future use
> + */
> +union sfh_event_base {
> +	u32 ul;
> +	struct {
> +		u32 response : 4;
> +		u32 status : 3;
> +		u32 out_in_c2p : 1;
> +		u32 length : 6;
> +		u32 dbg : 2;
> +		u32 sensor_id : 8;
> +		u32 rsvd : 8;
> +	} s;
> +};
> +
> +struct sfi_event_register {
> +	union sfh_event_base evtbase;
> +	u32 debuginfo1;
> +	u32 debuginfo2;
> +	u32 activecontrolstatus;
> +};
> +
> +enum sensor_idx {
> +	ACCEL_IDX               = 0,
> +	GYRO_IDX                = 1,
> +	MAG_IDX                 = 2,
> +	AMBIENT_LIGHT_IDX       = 19,
> +	NUM_ALL_SENSOR_CONSUMERS
> +};
> +
> +struct amd_mp2_dev {
> +	struct pci_dev *pdev;
> +	void __iomem *mmio;
> +	union sfh_event_base eventval;
> +	struct sfi_event_register eventreg;
> +	struct delayed_work work;
> +	void *ctx;
> +	void *cl_data;
> +};
> +
> +struct amd_mp2_sensor_info {
> +	u8 sensor_idx;
> +	u32 period;
> +	phys_addr_t phy_address;
> +};
> +
> +void amd_start_sensor(struct pci_dev *pdev, struct
> amd_mp2_sensor_info info);
> +void amd_stop_sensor(struct pci_dev *pdev, u16 sensor_idx);
> +void amd_stop_all_sensors(struct pci_dev *pdev);
> +int amd_mp2_get_sensor_num(struct pci_dev *dev, u8 *sensor_id);
> +#define ndev_pdev(ndev) ((ndev)->pdev)
> +#define ndev_name(ndev) pci_name(ndev_pdev(ndev))
> +#define ndev_dev(ndev) (&ndev_pdev(ndev)->dev)
> +#endif
Shah, Nehal-bakulchandra March 31, 2020, 1:18 p.m. UTC | #6
Hi Richard,

On 3/31/2020 6:01 PM, Richard Neumann wrote:
> Not a real review, but your patch series seems to be repeating a lot
> from drivers/i2c/busses/i2c-amd-mp2*.
> Is there any chance we could re-use the code?
> E.g. the AMD_C2P_* definitions from drivers/i2c/busses/i2c-amd-mp2.h?

Thanks for the mail. Yes there are some common structures, however as of now we have kept separately considering both

are part of different sub systems. But may be will consider this input for future enhancement.

Thanks

Nehal Shah

>
> Am Donnerstag, den 27.02.2020, 10:28 +0530 schrieb Sandeep Singh:
>> From: Sandeep Singh <sandeep.singh@amd.com>
>>
>> AMD SFH uses HID over PCIe bus.SFH fw is part of MP2
>> processor (MP2 which is an ARM® Cortex-M4 core based
>> co-processor to x86) and it runs on MP2 where in driver resides
>> on X86. This part of module will communicate with MP2 FW and
>> provide that data into DRAM
>>
>> Signed-off-by: Sandeep Singh <sandeep.singh@amd.com>
>> Signed-off-by: Nehal Shah <Nehal-bakulchandra.Shah@amd.com>
>> Reported-by: kbuild test robot <lkp@intel.com>
>> ---
>>  drivers/hid/Kconfig                    |   2 +
>>  drivers/hid/Makefile                   |   1 +
>>  drivers/hid/amd-sfh-hid/Kconfig        |  20 +++
>>  drivers/hid/amd-sfh-hid/Makefile       |  16 +++
>>  drivers/hid/amd-sfh-hid/amd_mp2_pcie.c | 243
>> +++++++++++++++++++++++++++++++++
>>  drivers/hid/amd-sfh-hid/amd_mp2_pcie.h | 176
>> ++++++++++++++++++++++++
>>  6 files changed, 458 insertions(+)
>>  create mode 100644 drivers/hid/amd-sfh-hid/Kconfig
>>  create mode 100644 drivers/hid/amd-sfh-hid/Makefile
>>  create mode 100644 drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
>>  create mode 100644 drivers/hid/amd-sfh-hid/amd_mp2_pcie.h
>>
>> diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig
>> index 494a39e..b253ad1 100644
>> --- a/drivers/hid/Kconfig
>> +++ b/drivers/hid/Kconfig
>> @@ -1155,4 +1155,6 @@ source "drivers/hid/i2c-hid/Kconfig"
>>  
>>  source "drivers/hid/intel-ish-hid/Kconfig"
>>  
>> +source "drivers/hid/amd-sfh-hid/Kconfig"
>> +
>>  endmenu
>> diff --git a/drivers/hid/Makefile b/drivers/hid/Makefile
>> index bfefa36..15a08e8 100644
>> --- a/drivers/hid/Makefile
>> +++ b/drivers/hid/Makefile
>> @@ -139,3 +139,4 @@ obj-$(CONFIG_I2C_HID)		+= i2c-hid/
>>  
>>  obj-$(CONFIG_INTEL_ISH_HID)	+= intel-ish-hid/
>>  obj-$(INTEL_ISH_FIRMWARE_DOWNLOADER)	+= intel-ish-hid/
>> +obj-$(CONFIG_AMD_SFH_HID)       += amd-sfh-hid/
>> diff --git a/drivers/hid/amd-sfh-hid/Kconfig b/drivers/hid/amd-sfh-
>> hid/Kconfig
>> new file mode 100644
>> index 0000000..7a224a1
>> --- /dev/null
>> +++ b/drivers/hid/amd-sfh-hid/Kconfig
>> @@ -0,0 +1,20 @@
>> +# SPDX-License-Identifier: GPL-2.0-only
>> +menu "AMD SFH HID support"
>> +	depends on (X86_64 || COMPILE_TEST) && PCI
>> +
>> +config AMD_SFH_HID
>> +	tristate "AMD Sensor Fusion Hub"
>> +	select HID
>> +	help
>> +	If you say yes to this option, support will be included for the
>> AMD
>> +	Sensor Fusion Hub.
>> +	This driver will enable sensors functionality to user through
>> HID
>> +	framework. Basically this driver will get data from MP2 FW
>> +	and provide that data to HID framework.
>> +	MP2 which is an ARM® Cortex-M4 core based co-processor to x86.
>> +
>> +	This driver can also be built as modules. If so, the modules
>> will
>> +	be  called amd-mp2-pcie and amd-sfhtp-hid.
>> +	Say Y or M here if you want to support AMD SFH. If unsure, say
>> N.
>> +
>> +endmenu
>> diff --git a/drivers/hid/amd-sfh-hid/Makefile b/drivers/hid/amd-sfh-
>> hid/Makefile
>> new file mode 100644
>> index 0000000..fa38d84
>> --- /dev/null
>> +++ b/drivers/hid/amd-sfh-hid/Makefile
>> @@ -0,0 +1,16 @@
>> +# SPDX-License-Identifier: GPL-2.0
>> +#
>> +# Makefile - AMD SFH HID drivers
>> +# Copyright (c) 2020-2021, Advanced Micro Devices, Inc.
>> +#
>> +#
>> +ccflags-m := -Werror
>> +obj-$(CONFIG_AMD_SFH_HID) += amd-mp2-pcie.o
>> +amd-mp2-pcie-objs := amd_mp2_pcie.o
>> +
>> +obj-$(CONFIG_AMD_SFH_HID) +=amd-sfhtp-hid.o
>> +amd-sfhtp-hid-objs := amdsfh-hid.o
>> +amd-sfhtp-hid-objs+= amdsfh-hid-client.o
>> +amd-sfhtp-hid-objs+= hid_descriptor/amd_sfh_hid_descriptor.o
>> +
>> +ccflags-y += -I$(srctree)/$(src)/
>> diff --git a/drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
>> b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
>> new file mode 100644
>> index 0000000..c67f389
>> --- /dev/null
>> +++ b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
>> @@ -0,0 +1,243 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +/*
>> + * AMD MP2 PCIe communication driver
>> + *
>> + * Authors: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
>> + *          Nehal Bakulchandra Shah <Nehal-bakulchandra.Shah@amd.com
>> + */
>> +
>> +#include <linux/interrupt.h>
>> +#include <linux/module.h>
>> +#include <linux/pci.h>
>> +#include <linux/slab.h>
>> +#include <linux/delay.h>
>> +#include "amd_mp2_pcie.h"
>> +
>> +#define DRIVER_NAME	"pcie_mp2_amd"
>> +#define DRIVER_DESC	"AMD(R) PCIe MP2 Communication Driver"
>> +
>> +#define	ACEL_EN		BIT(ACCEL_IDX)
>> +#define	GYRO_EN		BIT(GYRO_IDX)
>> +#define MAGNO_EN	BIT(MAG_IDX)
>> +#define ALS_EN		BIT(AMBIENT_LIGHT_IDX)
>> +
>> +void amd_start_sensor(struct pci_dev *pdev, struct
>> amd_mp2_sensor_info info)
>> +{
>> +	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
>> +	union sfh_command_parameter cmd_param;
>> +	union sfh_cmd_base cmd_base;
>> +
>> +	/*fill up command register*/
>> +	cmd_base.ul = 0;
>> +	cmd_base.s.cmd_id = enable_sensor;
>> +	cmd_base.s.period = info.period;
>> +	cmd_base.s.sensor_id = info.sensor_idx;
>> +
>> +	/*fill up command param register*/
>> +	cmd_param.ul = 0;
>> +	cmd_param.s.buffer_layout = 1;
>> +	cmd_param.s.buffer_length = 16;
>> +
>> +	write64((u64)info.phy_address, privdata->mmio + AMD_C2P_MSG2);
>> +	writel(cmd_param.ul, privdata->mmio + AMD_C2P_MSG1);
>> +	writel(cmd_base.ul, privdata->mmio + AMD_C2P_MSG0);
>> +}
>> +EXPORT_SYMBOL_GPL(amd_start_sensor);
>> +
>> +void amd_stop_sensor(struct pci_dev *pdev, u16 sensor_idx)
>> +{
>> +	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
>> +	union sfh_cmd_base cmd_base;
>> +
>> +	/* fill up command register */
>> +	cmd_base.ul = 0;
>> +	cmd_base.s.cmd_id = disable_sensor;
>> +	cmd_base.s.period = 0;
>> +	cmd_base.s.sensor_id = sensor_idx;
>> +
>> +	write64(0x0, privdata->mmio + AMD_C2P_MSG2);
>> +	writel(cmd_base.ul, privdata->mmio + AMD_C2P_MSG0);
>> +}
>> +EXPORT_SYMBOL_GPL(amd_stop_sensor);
>> +
>> +void amd_stop_all_sensors(struct pci_dev *pdev)
>> +{
>> +	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
>> +	union sfh_cmd_base cmd_base;
>> +
>> +	/*fill up command register */
>> +	cmd_base.ul = 0;
>> +	cmd_base.s.cmd_id = stop_all_sensors;
>> +	cmd_base.s.period = 0;
>> +	cmd_base.s.sensor_id = 0;
>> +
>> +	writel(cmd_base.ul, privdata->mmio + AMD_C2P_MSG0);
>> +}
>> +EXPORT_SYMBOL_GPL(amd_stop_all_sensors);
>> +
>> +int amd_mp2_get_sensor_num(struct pci_dev *dev, u8 *sensor_id)
>> +{
>> +	struct amd_mp2_dev *privdata = pci_get_drvdata(dev);
>> +	int activestatus;
>> +	int num_of_sensors = 0;
>> +
>> +	if (!sensor_id)
>> +		return -ENOMEM;
>> +
>> +	privdata->eventreg.activecontrolstatus =
>> +			readl(privdata->mmio + AMD_P2C_MSG3);
>> +	activestatus = privdata->eventreg.activecontrolstatus >> 4;
>> +
>> +	if (ACEL_EN  & activestatus) {
>> +		sensor_id[num_of_sensors] = ACCEL_IDX;
>> +		num_of_sensors++;
>> +	}
>> +	if (GYRO_EN & activestatus) {
>> +		sensor_id[num_of_sensors] = GYRO_IDX;
>> +		num_of_sensors++;
>> +	}
>> +	if (MAGNO_EN & activestatus) {
>> +		sensor_id[num_of_sensors] = MAG_IDX;
>> +		num_of_sensors++;
>> +	}
>> +
>> +	if (ALS_EN & activestatus) {
>> +		sensor_id[num_of_sensors] = AMBIENT_LIGHT_IDX;
>> +		num_of_sensors++;
>> +	}
>> +
>> +	return num_of_sensors;
>> +}
>> +EXPORT_SYMBOL_GPL(amd_mp2_get_sensor_num);
>> +
>> +static int amd_mp2_pci_init(struct amd_mp2_dev *privdata, struct
>> pci_dev *pdev)
>> +{
>> +	int rc;
>> +	int bar_index = 2;
>> +	resource_size_t size, base;
>> +
>> +	pci_set_drvdata(pdev, privdata);
>> +
>> +	rc = pcim_enable_device(pdev);
>> +	if (rc)
>> +		goto err_pci_enable;
>> +
>> +	rc = pcim_iomap_regions(pdev, 1 >> 2, DRIVER_NAME);
>> +	if (rc)
>> +		goto err_pci_regions;
>> +
>> +	privdata->mmio = pcim_iomap_table(pdev)[2];
>> +	pci_set_master(pdev);
>> +
>> +	rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
>> +	if (rc) {
>> +		rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
>> +		if (rc)
>> +			goto err_dma_mask;
>> +	}
>> +
>> +	rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
>> +	if (rc) {
>> +		rc = pci_set_consistent_dma_mask(pdev,
>> DMA_BIT_MASK(32));
>> +		if (rc)
>> +			goto err_dma_mask;
>> +	}
>> +
>> +	base = pci_resource_start(pdev, bar_index);
>> +	size = pci_resource_len(pdev, bar_index);
>> +	dev_dbg(ndev_dev(privdata), "Base addr:%llx size:%llx\n",
>> +		(unsigned long long)base, (unsigned long long)size);
>> +
>> +	privdata->mmio = ioremap(base, size);
>> +	if (!privdata->mmio) {
>> +		rc = -EIO;
>> +		goto err_dma_mask;
>> +	}
>> +
>> +	return 0;
>> +
>> +err_dma_mask:
>> +	pci_clear_master(pdev);
>> +err_pci_regions:
>> +	pci_disable_device(pdev);
>> +err_pci_enable:
>> +	pci_set_drvdata(pdev, NULL);
>> +	return rc;
>> +}
>> +
>> +static void amd_mp2_pci_deinit(struct amd_mp2_dev *privdata)
>> +{
>> +	struct pci_dev *pdev = ndev_pdev(privdata);
>> +
>> +	amd_stop_all_sensors(pdev);
>> +	pci_iounmap(pdev, privdata->mmio);
>> +
>> +	pci_clear_master(pdev);
>> +	pci_disable_device(pdev);
>> +	pci_set_drvdata(pdev, NULL);
>> +}
>> +
>> +static int amd_mp2_pci_probe(struct pci_dev *pdev,
>> +			     const struct pci_device_id *id)
>> +{
>> +	struct amd_mp2_dev *privdata;
>> +	int rc;
>> +
>> +	dev_info(&pdev->dev, "MP2 device found [%04x:%04x] (rev %x)\n",
>> +		 (int)pdev->vendor, (int)pdev->device, (int)pdev-
>>> revision);
>> +
>> +	privdata = devm_kzalloc(&pdev->dev, sizeof(*privdata),
>> GFP_KERNEL);
>> +
>> +	if (!privdata) {
>> +		rc = -ENOMEM;
>> +		goto err_dev;
>> +	}
>> +
>> +	privdata->pdev = pdev;
>> +
>> +	rc = amd_mp2_pci_init(privdata, pdev);
>> +	if (rc)
>> +		goto err_pci_init;
>> +	return 0;
>> +
>> +err_pci_init:
>> +	return rc;
>> +err_dev:
>> +	return rc;
>> +}
>> +
>> +static void amd_mp2_pci_remove(struct pci_dev *pdev)
>> +{
>> +	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
>> +
>> +	amd_mp2_pci_deinit(privdata);
>> +}
>> +
>> +static const struct pci_device_id amd_mp2_pci_tbl[] = {
>> +	{PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_MP2)},
>> +	{0}
>> +};
>> +MODULE_DEVICE_TABLE(pci, amd_mp2_pci_tbl);
>> +
>> +static struct pci_driver amd_mp2_pci_driver = {
>> +	.name		= DRIVER_NAME,
>> +	.id_table	= amd_mp2_pci_tbl,
>> +	.probe		= amd_mp2_pci_probe,
>> +	.remove		= amd_mp2_pci_remove,
>> +};
>> +
>> +static int __init amd_mp2_pci_driver_init(void)
>> +{
>> +	return pci_register_driver(&amd_mp2_pci_driver);
>> +}
>> +module_init(amd_mp2_pci_driver_init);
>> +
>> +static void __exit amd_mp2_pci_driver_exit(void)
>> +{
>> +	pci_unregister_driver(&amd_mp2_pci_driver);
>> +}
>> +module_exit(amd_mp2_pci_driver_exit);
>> +MODULE_DESCRIPTION(DRIVER_DESC);
>> +MODULE_LICENSE("Dual BSD/GPL");
>> +MODULE_AUTHOR("Shyam Sundar S K <Shyam-sundar.S-k@amd.com>");
>> +MODULE_AUTHOR("Nehal Bakulchandra Shah <
>> Nehal-bakulchandra.Shah@amd.com>");
>> diff --git a/drivers/hid/amd-sfh-hid/amd_mp2_pcie.h
>> b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.h
>> new file mode 100644
>> index 0000000..3ba69ac
>> --- /dev/null
>> +++ b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.h
>> @@ -0,0 +1,176 @@
>> +/* SPDX-License-Identifier: GPL-2.0
>> + *
>> + * AMD MP2 PCIe communication driver
>> + *
>> + * Authors: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
>> + *          Nehal Bakulchandra Shah <Nehal-bakulchandra.Shah@amd.com
>> + */
>> +
>> +#ifndef PCIE_MP2_AMD_H
>> +#define PCIE_MP2_AMD_H
>> +
>> +#include <linux/pci.h>
>> +#define PCI_DEVICE_ID_AMD_MP2	0x15E4
>> +
>> +/* MP2 C2P Message Registers */
>> +#define AMD_C2P_MSG0	0x10500
>> +#define AMD_C2P_MSG1	0x10504
>> +#define AMD_C2P_MSG2	0x10508
>> +#define AMD_C2P_MSG3	0x1050c
>> +#define AMD_C2P_MSG4	0x10510
>> +#define AMD_C2P_MSG5	0x10514
>> +#define AMD_C2P_MSG6	0x10518
>> +#define AMD_C2P_MSG7	0x1051c
>> +#define AMD_C2P_MSG8	0x10520
>> +#define AMD_C2P_MSG9	0x10524
>> +
>> +/* MP2 P2C Message Registers */
>> +#define AMD_P2C_MSG0	0x10680 /*Do not use*/
>> +#define AMD_P2C_MSG1	0x10684
>> +#define AMD_P2C_MSG2	0x10688
>> +#define AMD_P2C_MSG3	0x1068C /*MP2 debug info*/
>> +#define AMD_P2C_MSG_INTEN	0x10690 /*MP2 int gen register*/
>> +#define AMD_P2C_MSG_INTSTS	0x10694 /*Interrupt sts*/
>> +
>> +#define write64 amdsfh_write64
>> +static inline void amdsfh_write64(u64 val, void __iomem *mmio)
>> +{
>> +	writel(val, mmio);
>> +	writel(val >> 32, mmio + sizeof(u32));
>> +}
>> +
>> +#define read64 amdsfh_read64
>> +static inline u64 amdsfh_read64(void __iomem *mmio)
>> +{
>> +	u64 low, high;
>> +
>> +	low = readl(mmio);
>> +	high = readl(mmio + sizeof(u32));
>> +	return low | (high << 32);
>> +}
>> +
>> +/*
>> + * SFH Command registers
>> + */
>> +union sfh_cmd_base {
>> +	u32 ul;
>> +	struct {
>> +		u32 cmd_id : 8;
>> +		u32 sensor_id : 8;
>> +		u32 period : 16;
>> +	} s;
>> +};
>> +
>> +union sfh_command_parameter {
>> +	u32 ul;
>> +	struct {
>> +		u32 buffer_layout : 2;
>> +		u32 buffer_length : 6;
>> +		u32 rsvd : 24;
>> +	} s;
>> +};
>> +
>> +struct sfh_command_register {
>> +	union sfh_cmd_base cmd_base;
>> +	union sfh_command_parameter cmd_param;
>> +	phys_addr_t phy_addr;
>> +};
>> +
>> +/*
>> + * SFH Response registers
>> + */
>> +enum response_type {
>> +	non_operationevent,
>> +	command_success,
>> +	command_failed,
>> +	sfi_dataready_event,
>> +	invalid_response = 0xff,
>> +};
>> +
>> +enum status_type {
>> +	cmd_success,
>> +	invalid_data_payload,
>> +	invalid_data_length,
>> +	invalid_sensor_id,
>> +	invalid_dram_addr,
>> +	invalid_command,
>> +	sensor_enabled,
>> +	sensor_disabled,
>> +	status_end,
>> +};
>> +
>> +enum command_id {
>> +	non_operation = 0,
>> +	enable_sensor = 1,
>> +	disable_sensor = 2,
>> +	dump_sensorinfo = 3,
>> +	numberof_sensordiscovered = 4,
>> +	who_am_i_regchipid = 5,
>> +	set_dcd_data = 6,
>> +	get_dcd_data = 7,
>> +	stop_all_sensors = 8,
>> +	invalid_cmd = 0xf,
>> +};
>> +
>> +/**
>> + * union sfh_event_base : bit access of C2P commands
>> + * @response: bit: 0..3 SFI response_type
>> + * @status: bit: 6..5 status_type
>> + * @out_in_c2p: bit: 5 0- output in DRAM,1-in C2PMsg
>> + * @length: bit: 8..13 length
>> + * @dbg:bit: 14.15 dbg msg include in p2c msg 1-2
>> + * @sensor_id:bit: 16..23 Sensor ID
>> + * @rsvd:bit: 24..31 Reservered for future use
>> + */
>> +union sfh_event_base {
>> +	u32 ul;
>> +	struct {
>> +		u32 response : 4;
>> +		u32 status : 3;
>> +		u32 out_in_c2p : 1;
>> +		u32 length : 6;
>> +		u32 dbg : 2;
>> +		u32 sensor_id : 8;
>> +		u32 rsvd : 8;
>> +	} s;
>> +};
>> +
>> +struct sfi_event_register {
>> +	union sfh_event_base evtbase;
>> +	u32 debuginfo1;
>> +	u32 debuginfo2;
>> +	u32 activecontrolstatus;
>> +};
>> +
>> +enum sensor_idx {
>> +	ACCEL_IDX               = 0,
>> +	GYRO_IDX                = 1,
>> +	MAG_IDX                 = 2,
>> +	AMBIENT_LIGHT_IDX       = 19,
>> +	NUM_ALL_SENSOR_CONSUMERS
>> +};
>> +
>> +struct amd_mp2_dev {
>> +	struct pci_dev *pdev;
>> +	void __iomem *mmio;
>> +	union sfh_event_base eventval;
>> +	struct sfi_event_register eventreg;
>> +	struct delayed_work work;
>> +	void *ctx;
>> +	void *cl_data;
>> +};
>> +
>> +struct amd_mp2_sensor_info {
>> +	u8 sensor_idx;
>> +	u32 period;
>> +	phys_addr_t phy_address;
>> +};
>> +
>> +void amd_start_sensor(struct pci_dev *pdev, struct
>> amd_mp2_sensor_info info);
>> +void amd_stop_sensor(struct pci_dev *pdev, u16 sensor_idx);
>> +void amd_stop_all_sensors(struct pci_dev *pdev);
>> +int amd_mp2_get_sensor_num(struct pci_dev *dev, u8 *sensor_id);
>> +#define ndev_pdev(ndev) ((ndev)->pdev)
>> +#define ndev_name(ndev) pci_name(ndev_pdev(ndev))
>> +#define ndev_dev(ndev) (&ndev_pdev(ndev)->dev)
>> +#endif
Andy Shevchenko March 31, 2020, 5:24 p.m. UTC | #7
On Tue, Mar 31, 2020 at 4:26 PM Shah, Nehal-bakulchandra
<nehal-bakulchandra.shah@amd.com> wrote:
> On 3/31/2020 6:01 PM, Richard Neumann wrote:
> > Not a real review, but your patch series seems to be repeating a lot
> > from drivers/i2c/busses/i2c-amd-mp2*.
> > Is there any chance we could re-use the code?
> > E.g. the AMD_C2P_* definitions from drivers/i2c/busses/i2c-amd-mp2.h?
>
> Thanks for the mail. Yes there are some common structures, however as of now we have kept separately considering both
>
> are part of different sub systems. But may be will consider this input for future enhancement.

It can be done in a form of shared definitions at least in
include/linux/platform_data/x86/amd-mp2.h or alike ...
Richard Neumann April 1, 2020, 4:28 p.m. UTC | #8
Am Dienstag, den 31.03.2020, 20:24 +0300 schrieb Andy Shevchenko:
> On Tue, Mar 31, 2020 at 4:26 PM Shah, Nehal-bakulchandra
> <nehal-bakulchandra.shah@amd.com> wrote:
> > On 3/31/2020 6:01 PM, Richard Neumann wrote:
> > > Not a real review, but your patch series seems to be repeating a
> > > lot
> > > from drivers/i2c/busses/i2c-amd-mp2*.
> > > Is there any chance we could re-use the code?
> > > E.g. the AMD_C2P_* definitions from drivers/i2c/busses/i2c-amd-
> > > mp2.h?
> > 
> > Thanks for the mail. Yes there are some common structures, however
> > as of now we have kept separately considering both
> > 
> > are part of different sub systems. But may be will consider this
> > input for future enhancement.
> 
> It can be done in a form of shared definitions at least in
> include/linux/platform_data/x86/amd-mp2.h or alike ...
> 

I managed to add support for the AMD SFH PCI device to i2c-amd-mp2* and
outsourced the headers to include/linux/i2c-amd-mp2.h. [1]
I also refactored the patch series (excluded the documentation) [2] to
use the PCI device now provided by i2c_amd_mp2_pci and removed some
duplicate and unncessary code.
The driver now consist of just one module (amd_sfhtp_hid).
Unfortunately I was not able to solve the problem, that I get AMD-Vi
IO_PAGE_FAULT errors when not booted with amd_iommu=off.

[1] https://gist.githubusercontent.com/conqp/4d726f86da8a8397d6e70091a124de67/raw/f97e88a0b44d98bfa1258cb73c8afe4dce7afa87/i2c-amd-mp2.patch
[2] https://gist.githubusercontent.com/conqp/67036e690aca89d08b958971edac283d/raw/2a1ef122f9c8c8e07164b6d597962ce7bbad6d45/amd-sfhtp.patch
Shah, Nehal-bakulchandra April 6, 2020, 5:26 a.m. UTC | #9
HI

On 4/1/2020 9:58 PM, Richard Neumann wrote:
> Am Dienstag, den 31.03.2020, 20:24 +0300 schrieb Andy Shevchenko:
>> On Tue, Mar 31, 2020 at 4:26 PM Shah, Nehal-bakulchandra
>> <nehal-bakulchandra.shah@amd.com> wrote:
>>> On 3/31/2020 6:01 PM, Richard Neumann wrote:
>>>> Not a real review, but your patch series seems to be repeating a
>>>> lot
>>>> from drivers/i2c/busses/i2c-amd-mp2*.
>>>> Is there any chance we could re-use the code?
>>>> E.g. the AMD_C2P_* definitions from drivers/i2c/busses/i2c-amd-
>>>> mp2.h?
>>> Thanks for the mail. Yes there are some common structures, however
>>> as of now we have kept separately considering both
>>>
>>> are part of different sub systems. But may be will consider this
>>> input for future enhancement.
>> It can be done in a form of shared definitions at least in
>> include/linux/platform_data/x86/amd-mp2.h or alike ...
>>
> I managed to add support for the AMD SFH PCI device to i2c-amd-mp2* and
> outsourced the headers to include/linux/i2c-amd-mp2.h. [1]
> I also refactored the patch series (excluded the documentation) [2] to
> use the PCI device now provided by i2c_amd_mp2_pci and removed some
> duplicate and unncessary code.
> The driver now consist of just one module (amd_sfhtp_hid).
> Unfortunately I was not able to solve the problem, that I get AMD-Vi
> IO_PAGE_FAULT errors when not booted with amd_iommu=off.
>
> [1] https://gist.githubusercontent.com/conqp/4d726f86da8a8397d6e70091a124de67/raw/f97e88a0b44d98bfa1258cb73c8afe4dce7afa87/i2c-amd-mp2.patch
> [2] https://gist.githubusercontent.com/conqp/67036e690aca89d08b958971edac283d/raw/2a1ef122f9c8c8e07164b6d597962ce7bbad6d45/amd-sfhtp.patch

Thanks for the patch and appreciate your efforts. At this point of time, we would like to have our first patch for SFH to be upstreamed and dont want to complicate the

with two sub systems and maintainers. Surely will consider this input for future enhancement. Thanks for your understanding.

Thanks


Nehal
Richard Neumann April 13, 2020, 1:33 p.m. UTC | #10
Am Montag, den 06.04.2020, 10:56 +0530 schrieb Shah, Nehal-
bakulchandra:
> HI
> 
> On 4/1/2020 9:58 PM, Richard Neumann wrote:
> > Am Dienstag, den 31.03.2020, 20:24 +0300 schrieb Andy Shevchenko:
> > > On Tue, Mar 31, 2020 at 4:26 PM Shah, Nehal-bakulchandra
> > > <nehal-bakulchandra.shah@amd.com> wrote:
> > > > On 3/31/2020 6:01 PM, Richard Neumann wrote:
> > > > > Not a real review, but your patch series seems to be
> > > > > repeating a
> > > > > lot
> > > > > from drivers/i2c/busses/i2c-amd-mp2*.
> > > > > Is there any chance we could re-use the code?
> > > > > E.g. the AMD_C2P_* definitions from drivers/i2c/busses/i2c-
> > > > > amd-
> > > > > mp2.h?
> > > > Thanks for the mail. Yes there are some common structures,
> > > > however
> > > > as of now we have kept separately considering both
> > > > 
> > > > are part of different sub systems. But may be will consider
> > > > this
> > > > input for future enhancement.
> > > It can be done in a form of shared definitions at least in
> > > include/linux/platform_data/x86/amd-mp2.h or alike ...
> > > 
> > I managed to add support for the AMD SFH PCI device to i2c-amd-mp2* 
> > and
> > outsourced the headers to include/linux/i2c-amd-mp2.h. [1]
> > I also refactored the patch series (excluded the documentation) [2]
> > to
> > use the PCI device now provided by i2c_amd_mp2_pci and removed some
> > duplicate and unncessary code.
> > The driver now consist of just one module (amd_sfhtp_hid).
> > Unfortunately I was not able to solve the problem, that I get AMD-
> > Vi
> > IO_PAGE_FAULT errors when not booted with amd_iommu=off.
> > 
> > [1] 
> > https://gist.githubusercontent.com/conqp/4d726f86da8a8397d6e70091a124de67/raw/f97e88a0b44d98bfa1258cb73c8afe4dce7afa87/i2c-amd-mp2.patch
> > [2] 
> > https://gist.githubusercontent.com/conqp/67036e690aca89d08b958971edac283d/raw/2a1ef122f9c8c8e07164b6d597962ce7bbad6d45/amd-sfhtp.patch
> 
> Thanks for the patch and appreciate your efforts. At this point of
> time, we would like to have our first patch for SFH to be upstreamed
> and dont want to complicate the
> 
> with two sub systems and maintainers. Surely will consider this input
> for future enhancement. Thanks for your understanding.
> 
> Thanks
> 
> 
> Nehal
> 
> 
Hi all,

I spent the last two weeks learning the HID API and refactored [1] the
patch series again. I also took into consideration Nehal's input
regarding the two subsystems. After tinkering with the integration of
the SFH PCI driver into the existing i2c-amd-mp2-pci driver, I realized
that this was a futile effort at this point of time, since the SFH
protocol, as far as I understand it, does not really implement the I2C
protocol. Thusly I turned away from trying to integrate the two
drivers. The core changes I did, is to refactor the PCI driver to use a
similar initialization as the i2c-amd-mp2-pci driver. The registers are
now mapped correctly on initial setup (1 << 2) and not mapped to zero
(1 >> 2) and the ioremapped later. With this change, the AMD-Vi page
faults when not booting with amd_iommu=off also automagically
disappeared. 
The second change is the HID device handling. I now implemented the
hid_ll_driver API with the respective functions to process raw events
as well device start / stop and open / close routines. I also
refactored the platform driver wich is now solely responsible for
spawning and despawning the HID devices. I also reduced the commonly
used structures and constats to those that were actually being used by
the driver and renamed the PCI registers to express what they are used
for in this driver. I also commented most functions and shared
structures. The only problem I keep having on my machine is, that the
register for the bit-shifted active sensors mask 0x1068C keeps
returning zero on readout and thusly I need to override it.
I tried to get the sensor mask by sending the "dump_sensorinfo" and
"number_of_sensors_discovered" commands to different C2P registers and
dumping the corresponding P2C registers. But so far I could not get any
information that could be interpreted as the sensor info and mask that
I need. Maybe Nehal or Sandeep could provide the information, if and
how it ts possible to get information about the amount, position and
type of the connected sensors in another way. I also found out, by
sweeping the SFH registers (from 0 to 31), that there seem to be other
sensors connected on my machine, namely at positions 0, 1, 2, 3, 4, 7,
8 and 9. The ALS at 19 does not seem to be present on my system. I
suspect that at least one of those is an additional accelerometer,
since the only one working right now ist the accelerometer at position
0, which is located in the sceen / lid. I suspect that the other one is
located in the chassis, allowing the use of both to perform tablet mode
detection by processing their relative orientation. I'd appreciate a
hint on this assumption as well. ;-)

Sincere regards,

Richard

[1] https://gist.github.com/conqp/33baa079d9524914c4c0c196200e4f89
Shah, Nehal-bakulchandra April 21, 2020, 6:31 p.m. UTC | #11
Hi Richard,

On 4/13/2020 7:03 PM, Richard Neumann wrote:
> Am Montag, den 06.04.2020, 10:56 +0530 schrieb Shah, Nehal-
> bakulchandra:
>> HI
>>
>> On 4/1/2020 9:58 PM, Richard Neumann wrote:
>>> Am Dienstag, den 31.03.2020, 20:24 +0300 schrieb Andy Shevchenko:
>>>> On Tue, Mar 31, 2020 at 4:26 PM Shah, Nehal-bakulchandra
>>>> <nehal-bakulchandra.shah@amd.com> wrote:
>>>>> On 3/31/2020 6:01 PM, Richard Neumann wrote:
>>>>>> Not a real review, but your patch series seems to be
>>>>>> repeating a
>>>>>> lot
>>>>>> from drivers/i2c/busses/i2c-amd-mp2*.
>>>>>> Is there any chance we could re-use the code?
>>>>>> E.g. the AMD_C2P_* definitions from drivers/i2c/busses/i2c-
>>>>>> amd-
>>>>>> mp2.h?
>>>>> Thanks for the mail. Yes there are some common structures,
>>>>> however
>>>>> as of now we have kept separately considering both
>>>>>
>>>>> are part of different sub systems. But may be will consider
>>>>> this
>>>>> input for future enhancement.
>>>> It can be done in a form of shared definitions at least in
>>>> include/linux/platform_data/x86/amd-mp2.h or alike ...
>>>>
>>> I managed to add support for the AMD SFH PCI device to i2c-amd-mp2* 
>>> and
>>> outsourced the headers to include/linux/i2c-amd-mp2.h. [1]
>>> I also refactored the patch series (excluded the documentation) [2]
>>> to
>>> use the PCI device now provided by i2c_amd_mp2_pci and removed some
>>> duplicate and unncessary code.
>>> The driver now consist of just one module (amd_sfhtp_hid).
>>> Unfortunately I was not able to solve the problem, that I get AMD-
>>> Vi
>>> IO_PAGE_FAULT errors when not booted with amd_iommu=off.
>>>
>>> [1] 
>>> https://gist.githubusercontent.com/conqp/4d726f86da8a8397d6e70091a124de67/raw/f97e88a0b44d98bfa1258cb73c8afe4dce7afa87/i2c-amd-mp2.patch
>>> [2] 
>>> https://gist.githubusercontent.com/conqp/67036e690aca89d08b958971edac283d/raw/2a1ef122f9c8c8e07164b6d597962ce7bbad6d45/amd-sfhtp.patch
>> Thanks for the patch and appreciate your efforts. At this point of
>> time, we would like to have our first patch for SFH to be upstreamed
>> and dont want to complicate the
>>
>> with two sub systems and maintainers. Surely will consider this input
>> for future enhancement. Thanks for your understanding.
>>
>> Thanks
>>
>>
>> Nehal
>>
>>
> Hi all,
>
> I spent the last two weeks learning the HID API and refactored [1] the
> patch series again. I also took into consideration Nehal's input
> regarding the two subsystems. After tinkering with the integration of
> the SFH PCI driver into the existing i2c-amd-mp2-pci driver, I realized
> that this was a futile effort at this point of time, since the SFH
> protocol, as far as I understand it, does not really implement the I2C
> protocol. Thusly I turned away from trying to integrate the two
> drivers. The core changes I did, is to refactor the PCI driver to use a
> similar initialization as the i2c-amd-mp2-pci driver. The registers are
> now mapped correctly on initial setup (1 << 2) and not mapped to zero
> (1 >> 2) and the ioremapped later. With this change, the AMD-Vi page
> faults when not booting with amd_iommu=off also automagically
> disappeared. 
> The second change is the HID device handling. I now implemented the
> hid_ll_driver API with the respective functions to process raw events
> as well device start / stop and open / close routines. I also
> refactored the platform driver wich is now solely responsible for
> spawning and despawning the HID devices. I also reduced the commonly
> used structures and constats to those that were actually being used by
> the driver and renamed the PCI registers to express what they are used
> for in this driver. I also commented most functions and shared
> structures. The only problem I keep having on my machine is, that the
> register for the bit-shifted active sensors mask 0x1068C keeps
> returning zero on readout and thusly I need to override it.
> I tried to get the sensor mask by sending the "dump_sensorinfo" and
> "number_of_sensors_discovered" commands to different C2P registers and
> dumping the corresponding P2C registers. But so far I could not get any
> information that could be interpreted as the sensor info and mask that
> I need. Maybe Nehal or Sandeep could provide the information, if and
> how it ts possible to get information about the amount, position and
> type of the connected sensors in another way. I also found out, by
> sweeping the SFH registers (from 0 to 31), that there seem to be other
> sensors connected on my machine, namely at positions 0, 1, 2, 3, 4, 7,
> 8 and 9. The ALS at 19 does not seem to be present on my system. I
> suspect that at least one of those is an additional accelerometer,
> since the only one working right now ist the accelerometer at position
> 0, which is located in the sceen / lid. I suspect that the other one is
> located in the chassis, allowing the use of both to perform tablet mode
> detection by processing their relative orientation. I'd appreciate a
> hint on this assumption as well. ;-)
>
> Sincere regards,
>
> Richard
>
> [1] https://gist.github.com/conqp/33baa079d9524914c4c0c196200e4f89

Thanks for the refactoring  the patch. The .raw_request in hid_ll_driver is not correct, the output of sensor is not raw but it is processed one,

so better to move it to .request function. Regarding your question for sensor position well it comes from firmware. So i am not sure about the firmware

for sensor , what you have.Also, regarding the IOMMU, we figured out the issue, earlier patch series has bug that during DMA allocation pdev was passed

from platform driver context, where in for IOMMU needs context PCI pdev. So that is taken care in your refactored patch hence issue is disappeared.


Now regarding the patch (with your changes) i need to validate at our side. Due to lockdown i dont have access to the system, so may be we can wait on that.


Thanks

Nehal Shah
Richard Neumann April 21, 2020, 9:18 p.m. UTC | #12
Am Mittwoch, den 22.04.2020, 00:01 +0530 schrieb Shah, Nehal-
bakulchandra:
> Hi Richard,
> 
> Thanks for the refactoring  the patch. The .raw_request in
> hid_ll_driver is not correct, the output of sensor is not raw but it
> is processed one,
> 
> so better to move it to .request function. Regarding your question
> for sensor position well it comes from firmware. So i am not sure
> about the firmware
> 
> for sensor , what you have.Also, regarding the IOMMU, we figured out
> the issue, earlier patch series has bug that during DMA allocation
> pdev was passed
> 
> from platform driver context, where in for IOMMU needs context PCI
> pdev. So that is taken care in your refactored patch hence issue is
> disappeared.
> 
> 
> Now regarding the patch (with your changes) i need to validate at our
> side. Due to lockdown i dont have access to the system, so may be we
> can wait on that.
> 
> 
> Thanks
> 
> Nehal Shah

Hi Nehal

and thank you for the feedback.

Regarding the raw_request / request, I think the "raw" here is meant to
indicate, that the implementation is expected to write "raw bytes" [1]
into a buffer rather than handling a "struct hid_report", which is
exactly what the get_feature_report() and get_input_report() from the
original patch seem to be intended to do.
However it should be easy to migrate this to ".report", since it'd
probably look a lot like __hid_request() [2], which is currently doing
the magic automatically for me.
Furthermore a ll_driver needs a raw_request implementation anyways as
it's being tested for in hid_add_device() [3].

Regarding the validation on your side: Of course! I do in no way want
to undermine your work on this in any way. I'm just offering what I've
learned so far and what is working for me. If you find parts of it
worth taking into a refactored version for upstream, awesome. If you
don't agree with other parts, that's fine, too. I'm quite new to this
and eager to learn. Also I am in no rush. Good things take time.

In case you're missing my patch on gist.github.com, I decided to
further develop the driver in a forked linux source tree [4].

It's on the branch amd-sfh. You can just diff to the master branch to
get the entire patch. The changed files are still under
drivers/hid/amd-sfh-hid and Documentation/hid respectively.

You'll find that in the meantime I did some more work / learning and
implemented some rudimentary IRQ handling.

Kind regards,

Richard

[1] https://elixir.bootlin.com/linux/latest/source/include/linux/hid.h#L1070
[2] https://elixir.bootlin.com/linux/latest/source/drivers/hid/hid-core.c#L1688
[3] https://elixir.bootlin.com/linux/latest/source/drivers/hid/hid-core.c#L2386
[3] https://github.com/conqp/linux/tree/amd-sfh
diff mbox series

Patch

diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig
index 494a39e..b253ad1 100644
--- a/drivers/hid/Kconfig
+++ b/drivers/hid/Kconfig
@@ -1155,4 +1155,6 @@  source "drivers/hid/i2c-hid/Kconfig"
 
 source "drivers/hid/intel-ish-hid/Kconfig"
 
+source "drivers/hid/amd-sfh-hid/Kconfig"
+
 endmenu
diff --git a/drivers/hid/Makefile b/drivers/hid/Makefile
index bfefa36..15a08e8 100644
--- a/drivers/hid/Makefile
+++ b/drivers/hid/Makefile
@@ -139,3 +139,4 @@  obj-$(CONFIG_I2C_HID)		+= i2c-hid/
 
 obj-$(CONFIG_INTEL_ISH_HID)	+= intel-ish-hid/
 obj-$(INTEL_ISH_FIRMWARE_DOWNLOADER)	+= intel-ish-hid/
+obj-$(CONFIG_AMD_SFH_HID)       += amd-sfh-hid/
diff --git a/drivers/hid/amd-sfh-hid/Kconfig b/drivers/hid/amd-sfh-hid/Kconfig
new file mode 100644
index 0000000..7a224a1
--- /dev/null
+++ b/drivers/hid/amd-sfh-hid/Kconfig
@@ -0,0 +1,20 @@ 
+# SPDX-License-Identifier: GPL-2.0-only
+menu "AMD SFH HID support"
+	depends on (X86_64 || COMPILE_TEST) && PCI
+
+config AMD_SFH_HID
+	tristate "AMD Sensor Fusion Hub"
+	select HID
+	help
+	If you say yes to this option, support will be included for the AMD
+	Sensor Fusion Hub.
+	This driver will enable sensors functionality to user through HID
+	framework. Basically this driver will get data from MP2 FW
+	and provide that data to HID framework.
+	MP2 which is an ARM® Cortex-M4 core based co-processor to x86.
+
+	This driver can also be built as modules. If so, the modules will
+	be  called amd-mp2-pcie and amd-sfhtp-hid.
+	Say Y or M here if you want to support AMD SFH. If unsure, say N.
+
+endmenu
diff --git a/drivers/hid/amd-sfh-hid/Makefile b/drivers/hid/amd-sfh-hid/Makefile
new file mode 100644
index 0000000..fa38d84
--- /dev/null
+++ b/drivers/hid/amd-sfh-hid/Makefile
@@ -0,0 +1,16 @@ 
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile - AMD SFH HID drivers
+# Copyright (c) 2020-2021, Advanced Micro Devices, Inc.
+#
+#
+ccflags-m := -Werror
+obj-$(CONFIG_AMD_SFH_HID) += amd-mp2-pcie.o
+amd-mp2-pcie-objs := amd_mp2_pcie.o
+
+obj-$(CONFIG_AMD_SFH_HID) +=amd-sfhtp-hid.o
+amd-sfhtp-hid-objs := amdsfh-hid.o
+amd-sfhtp-hid-objs+= amdsfh-hid-client.o
+amd-sfhtp-hid-objs+= hid_descriptor/amd_sfh_hid_descriptor.o
+
+ccflags-y += -I$(srctree)/$(src)/
diff --git a/drivers/hid/amd-sfh-hid/amd_mp2_pcie.c b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
new file mode 100644
index 0000000..c67f389
--- /dev/null
+++ b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.c
@@ -0,0 +1,243 @@ 
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * AMD MP2 PCIe communication driver
+ *
+ * Authors: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
+ *          Nehal Bakulchandra Shah <Nehal-bakulchandra.Shah@amd.com>
+ */
+
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include "amd_mp2_pcie.h"
+
+#define DRIVER_NAME	"pcie_mp2_amd"
+#define DRIVER_DESC	"AMD(R) PCIe MP2 Communication Driver"
+
+#define	ACEL_EN		BIT(ACCEL_IDX)
+#define	GYRO_EN		BIT(GYRO_IDX)
+#define MAGNO_EN	BIT(MAG_IDX)
+#define ALS_EN		BIT(AMBIENT_LIGHT_IDX)
+
+void amd_start_sensor(struct pci_dev *pdev, struct amd_mp2_sensor_info info)
+{
+	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
+	union sfh_command_parameter cmd_param;
+	union sfh_cmd_base cmd_base;
+
+	/*fill up command register*/
+	cmd_base.ul = 0;
+	cmd_base.s.cmd_id = enable_sensor;
+	cmd_base.s.period = info.period;
+	cmd_base.s.sensor_id = info.sensor_idx;
+
+	/*fill up command param register*/
+	cmd_param.ul = 0;
+	cmd_param.s.buffer_layout = 1;
+	cmd_param.s.buffer_length = 16;
+
+	write64((u64)info.phy_address, privdata->mmio + AMD_C2P_MSG2);
+	writel(cmd_param.ul, privdata->mmio + AMD_C2P_MSG1);
+	writel(cmd_base.ul, privdata->mmio + AMD_C2P_MSG0);
+}
+EXPORT_SYMBOL_GPL(amd_start_sensor);
+
+void amd_stop_sensor(struct pci_dev *pdev, u16 sensor_idx)
+{
+	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
+	union sfh_cmd_base cmd_base;
+
+	/* fill up command register */
+	cmd_base.ul = 0;
+	cmd_base.s.cmd_id = disable_sensor;
+	cmd_base.s.period = 0;
+	cmd_base.s.sensor_id = sensor_idx;
+
+	write64(0x0, privdata->mmio + AMD_C2P_MSG2);
+	writel(cmd_base.ul, privdata->mmio + AMD_C2P_MSG0);
+}
+EXPORT_SYMBOL_GPL(amd_stop_sensor);
+
+void amd_stop_all_sensors(struct pci_dev *pdev)
+{
+	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
+	union sfh_cmd_base cmd_base;
+
+	/*fill up command register */
+	cmd_base.ul = 0;
+	cmd_base.s.cmd_id = stop_all_sensors;
+	cmd_base.s.period = 0;
+	cmd_base.s.sensor_id = 0;
+
+	writel(cmd_base.ul, privdata->mmio + AMD_C2P_MSG0);
+}
+EXPORT_SYMBOL_GPL(amd_stop_all_sensors);
+
+int amd_mp2_get_sensor_num(struct pci_dev *dev, u8 *sensor_id)
+{
+	struct amd_mp2_dev *privdata = pci_get_drvdata(dev);
+	int activestatus;
+	int num_of_sensors = 0;
+
+	if (!sensor_id)
+		return -ENOMEM;
+
+	privdata->eventreg.activecontrolstatus =
+			readl(privdata->mmio + AMD_P2C_MSG3);
+	activestatus = privdata->eventreg.activecontrolstatus >> 4;
+
+	if (ACEL_EN  & activestatus) {
+		sensor_id[num_of_sensors] = ACCEL_IDX;
+		num_of_sensors++;
+	}
+	if (GYRO_EN & activestatus) {
+		sensor_id[num_of_sensors] = GYRO_IDX;
+		num_of_sensors++;
+	}
+	if (MAGNO_EN & activestatus) {
+		sensor_id[num_of_sensors] = MAG_IDX;
+		num_of_sensors++;
+	}
+
+	if (ALS_EN & activestatus) {
+		sensor_id[num_of_sensors] = AMBIENT_LIGHT_IDX;
+		num_of_sensors++;
+	}
+
+	return num_of_sensors;
+}
+EXPORT_SYMBOL_GPL(amd_mp2_get_sensor_num);
+
+static int amd_mp2_pci_init(struct amd_mp2_dev *privdata, struct pci_dev *pdev)
+{
+	int rc;
+	int bar_index = 2;
+	resource_size_t size, base;
+
+	pci_set_drvdata(pdev, privdata);
+
+	rc = pcim_enable_device(pdev);
+	if (rc)
+		goto err_pci_enable;
+
+	rc = pcim_iomap_regions(pdev, 1 >> 2, DRIVER_NAME);
+	if (rc)
+		goto err_pci_regions;
+
+	privdata->mmio = pcim_iomap_table(pdev)[2];
+	pci_set_master(pdev);
+
+	rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
+	if (rc) {
+		rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
+		if (rc)
+			goto err_dma_mask;
+	}
+
+	rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
+	if (rc) {
+		rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32));
+		if (rc)
+			goto err_dma_mask;
+	}
+
+	base = pci_resource_start(pdev, bar_index);
+	size = pci_resource_len(pdev, bar_index);
+	dev_dbg(ndev_dev(privdata), "Base addr:%llx size:%llx\n",
+		(unsigned long long)base, (unsigned long long)size);
+
+	privdata->mmio = ioremap(base, size);
+	if (!privdata->mmio) {
+		rc = -EIO;
+		goto err_dma_mask;
+	}
+
+	return 0;
+
+err_dma_mask:
+	pci_clear_master(pdev);
+err_pci_regions:
+	pci_disable_device(pdev);
+err_pci_enable:
+	pci_set_drvdata(pdev, NULL);
+	return rc;
+}
+
+static void amd_mp2_pci_deinit(struct amd_mp2_dev *privdata)
+{
+	struct pci_dev *pdev = ndev_pdev(privdata);
+
+	amd_stop_all_sensors(pdev);
+	pci_iounmap(pdev, privdata->mmio);
+
+	pci_clear_master(pdev);
+	pci_disable_device(pdev);
+	pci_set_drvdata(pdev, NULL);
+}
+
+static int amd_mp2_pci_probe(struct pci_dev *pdev,
+			     const struct pci_device_id *id)
+{
+	struct amd_mp2_dev *privdata;
+	int rc;
+
+	dev_info(&pdev->dev, "MP2 device found [%04x:%04x] (rev %x)\n",
+		 (int)pdev->vendor, (int)pdev->device, (int)pdev->revision);
+
+	privdata = devm_kzalloc(&pdev->dev, sizeof(*privdata), GFP_KERNEL);
+
+	if (!privdata) {
+		rc = -ENOMEM;
+		goto err_dev;
+	}
+
+	privdata->pdev = pdev;
+
+	rc = amd_mp2_pci_init(privdata, pdev);
+	if (rc)
+		goto err_pci_init;
+	return 0;
+
+err_pci_init:
+	return rc;
+err_dev:
+	return rc;
+}
+
+static void amd_mp2_pci_remove(struct pci_dev *pdev)
+{
+	struct amd_mp2_dev *privdata = pci_get_drvdata(pdev);
+
+	amd_mp2_pci_deinit(privdata);
+}
+
+static const struct pci_device_id amd_mp2_pci_tbl[] = {
+	{PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_MP2)},
+	{0}
+};
+MODULE_DEVICE_TABLE(pci, amd_mp2_pci_tbl);
+
+static struct pci_driver amd_mp2_pci_driver = {
+	.name		= DRIVER_NAME,
+	.id_table	= amd_mp2_pci_tbl,
+	.probe		= amd_mp2_pci_probe,
+	.remove		= amd_mp2_pci_remove,
+};
+
+static int __init amd_mp2_pci_driver_init(void)
+{
+	return pci_register_driver(&amd_mp2_pci_driver);
+}
+module_init(amd_mp2_pci_driver_init);
+
+static void __exit amd_mp2_pci_driver_exit(void)
+{
+	pci_unregister_driver(&amd_mp2_pci_driver);
+}
+module_exit(amd_mp2_pci_driver_exit);
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_AUTHOR("Shyam Sundar S K <Shyam-sundar.S-k@amd.com>");
+MODULE_AUTHOR("Nehal Bakulchandra Shah <Nehal-bakulchandra.Shah@amd.com>");
diff --git a/drivers/hid/amd-sfh-hid/amd_mp2_pcie.h b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.h
new file mode 100644
index 0000000..3ba69ac
--- /dev/null
+++ b/drivers/hid/amd-sfh-hid/amd_mp2_pcie.h
@@ -0,0 +1,176 @@ 
+/* SPDX-License-Identifier: GPL-2.0
+ *
+ * AMD MP2 PCIe communication driver
+ *
+ * Authors: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
+ *          Nehal Bakulchandra Shah <Nehal-bakulchandra.Shah@amd.com>
+ */
+
+#ifndef PCIE_MP2_AMD_H
+#define PCIE_MP2_AMD_H
+
+#include <linux/pci.h>
+#define PCI_DEVICE_ID_AMD_MP2	0x15E4
+
+/* MP2 C2P Message Registers */
+#define AMD_C2P_MSG0	0x10500
+#define AMD_C2P_MSG1	0x10504
+#define AMD_C2P_MSG2	0x10508
+#define AMD_C2P_MSG3	0x1050c
+#define AMD_C2P_MSG4	0x10510
+#define AMD_C2P_MSG5	0x10514
+#define AMD_C2P_MSG6	0x10518
+#define AMD_C2P_MSG7	0x1051c
+#define AMD_C2P_MSG8	0x10520
+#define AMD_C2P_MSG9	0x10524
+
+/* MP2 P2C Message Registers */
+#define AMD_P2C_MSG0	0x10680 /*Do not use*/
+#define AMD_P2C_MSG1	0x10684
+#define AMD_P2C_MSG2	0x10688
+#define AMD_P2C_MSG3	0x1068C /*MP2 debug info*/
+#define AMD_P2C_MSG_INTEN	0x10690 /*MP2 int gen register*/
+#define AMD_P2C_MSG_INTSTS	0x10694 /*Interrupt sts*/
+
+#define write64 amdsfh_write64
+static inline void amdsfh_write64(u64 val, void __iomem *mmio)
+{
+	writel(val, mmio);
+	writel(val >> 32, mmio + sizeof(u32));
+}
+
+#define read64 amdsfh_read64
+static inline u64 amdsfh_read64(void __iomem *mmio)
+{
+	u64 low, high;
+
+	low = readl(mmio);
+	high = readl(mmio + sizeof(u32));
+	return low | (high << 32);
+}
+
+/*
+ * SFH Command registers
+ */
+union sfh_cmd_base {
+	u32 ul;
+	struct {
+		u32 cmd_id : 8;
+		u32 sensor_id : 8;
+		u32 period : 16;
+	} s;
+};
+
+union sfh_command_parameter {
+	u32 ul;
+	struct {
+		u32 buffer_layout : 2;
+		u32 buffer_length : 6;
+		u32 rsvd : 24;
+	} s;
+};
+
+struct sfh_command_register {
+	union sfh_cmd_base cmd_base;
+	union sfh_command_parameter cmd_param;
+	phys_addr_t phy_addr;
+};
+
+/*
+ * SFH Response registers
+ */
+enum response_type {
+	non_operationevent,
+	command_success,
+	command_failed,
+	sfi_dataready_event,
+	invalid_response = 0xff,
+};
+
+enum status_type {
+	cmd_success,
+	invalid_data_payload,
+	invalid_data_length,
+	invalid_sensor_id,
+	invalid_dram_addr,
+	invalid_command,
+	sensor_enabled,
+	sensor_disabled,
+	status_end,
+};
+
+enum command_id {
+	non_operation = 0,
+	enable_sensor = 1,
+	disable_sensor = 2,
+	dump_sensorinfo = 3,
+	numberof_sensordiscovered = 4,
+	who_am_i_regchipid = 5,
+	set_dcd_data = 6,
+	get_dcd_data = 7,
+	stop_all_sensors = 8,
+	invalid_cmd = 0xf,
+};
+
+/**
+ * union sfh_event_base : bit access of C2P commands
+ * @response: bit: 0..3 SFI response_type
+ * @status: bit: 6..5 status_type
+ * @out_in_c2p: bit: 5 0- output in DRAM,1-in C2PMsg
+ * @length: bit: 8..13 length
+ * @dbg:bit: 14.15 dbg msg include in p2c msg 1-2
+ * @sensor_id:bit: 16..23 Sensor ID
+ * @rsvd:bit: 24..31 Reservered for future use
+ */
+union sfh_event_base {
+	u32 ul;
+	struct {
+		u32 response : 4;
+		u32 status : 3;
+		u32 out_in_c2p : 1;
+		u32 length : 6;
+		u32 dbg : 2;
+		u32 sensor_id : 8;
+		u32 rsvd : 8;
+	} s;
+};
+
+struct sfi_event_register {
+	union sfh_event_base evtbase;
+	u32 debuginfo1;
+	u32 debuginfo2;
+	u32 activecontrolstatus;
+};
+
+enum sensor_idx {
+	ACCEL_IDX               = 0,
+	GYRO_IDX                = 1,
+	MAG_IDX                 = 2,
+	AMBIENT_LIGHT_IDX       = 19,
+	NUM_ALL_SENSOR_CONSUMERS
+};
+
+struct amd_mp2_dev {
+	struct pci_dev *pdev;
+	void __iomem *mmio;
+	union sfh_event_base eventval;
+	struct sfi_event_register eventreg;
+	struct delayed_work work;
+	void *ctx;
+	void *cl_data;
+};
+
+struct amd_mp2_sensor_info {
+	u8 sensor_idx;
+	u32 period;
+	phys_addr_t phy_address;
+};
+
+void amd_start_sensor(struct pci_dev *pdev, struct amd_mp2_sensor_info info);
+void amd_stop_sensor(struct pci_dev *pdev, u16 sensor_idx);
+void amd_stop_all_sensors(struct pci_dev *pdev);
+int amd_mp2_get_sensor_num(struct pci_dev *dev, u8 *sensor_id);
+#define ndev_pdev(ndev) ((ndev)->pdev)
+#define ndev_name(ndev) pci_name(ndev_pdev(ndev))
+#define ndev_dev(ndev) (&ndev_pdev(ndev)->dev)
+#endif