2016-10-10 13:48:15

by Benjamin Tissoires

[permalink] [raw]
Subject: Re: [PATCH v3 2/2] Input: synaptics-rmi4 - add support for F34 device reflash

On Sep 20 2016 or thereabouts, Nick Dyer wrote:
> Signed-off-by: Nick Dyer <[email protected]>
> Tested-by: Chris Healy <[email protected]>
> ---
> drivers/input/rmi4/Kconfig | 11 +
> drivers/input/rmi4/Makefile | 1 +
> drivers/input/rmi4/rmi_bus.c | 3 +
> drivers/input/rmi4/rmi_driver.c | 161 ++++++++++++++-
> drivers/input/rmi4/rmi_driver.h | 7 +
> drivers/input/rmi4/rmi_f01.c | 6 +
> drivers/input/rmi4/rmi_f34.c | 446 ++++++++++++++++++++++++++++++++++++++++
> include/linux/rmi.h | 1 +
> 8 files changed, 634 insertions(+), 2 deletions(-)
> create mode 100644 drivers/input/rmi4/rmi_f34.c
>
> diff --git a/drivers/input/rmi4/Kconfig b/drivers/input/rmi4/Kconfig
> index f73df24..24c3e1b 100644
> --- a/drivers/input/rmi4/Kconfig
> +++ b/drivers/input/rmi4/Kconfig
> @@ -61,3 +61,14 @@ config RMI4_F30
>
> Function 30 provides GPIO and LED support for RMI4 devices. This
> includes support for buttons on TouchPads and ClickPads.
> +
> +config RMI4_F34
> + bool "RMI4 Function 34 (Device reflash)"
> + depends on RMI4_CORE
> + select FW_LOADER
> + help
> + Say Y here if you want to add support for RMI4 function 34.
> +
> + Function 34 provides support for upgrading the firmware on the RMI4
> + device via the firmware loader interface. This is triggered using a
> + sysfs attribute.
> diff --git a/drivers/input/rmi4/Makefile b/drivers/input/rmi4/Makefile
> index 95c00a7..54e6bfe 100644
> --- a/drivers/input/rmi4/Makefile
> +++ b/drivers/input/rmi4/Makefile
> @@ -7,6 +7,7 @@ rmi_core-$(CONFIG_RMI4_2D_SENSOR) += rmi_2d_sensor.o
> rmi_core-$(CONFIG_RMI4_F11) += rmi_f11.o
> rmi_core-$(CONFIG_RMI4_F12) += rmi_f12.o
> rmi_core-$(CONFIG_RMI4_F30) += rmi_f30.o
> +rmi_core-$(CONFIG_RMI4_F34) += rmi_f34.o
>
> # Transports
> obj-$(CONFIG_RMI4_I2C) += rmi_i2c.o
> diff --git a/drivers/input/rmi4/rmi_bus.c b/drivers/input/rmi4/rmi_bus.c
> index a735806..be0aea9 100644
> --- a/drivers/input/rmi4/rmi_bus.c
> +++ b/drivers/input/rmi4/rmi_bus.c
> @@ -312,6 +312,9 @@ static struct rmi_function_handler *fn_handlers[] = {
> #ifdef CONFIG_RMI4_F30
> &rmi_f30_handler,
> #endif
> +#ifdef CONFIG_RMI4_F34
> + &rmi_f34_handler,
> +#endif
> };
>
> static void __rmi_unregister_function_handlers(int start_idx)
> diff --git a/drivers/input/rmi4/rmi_driver.c b/drivers/input/rmi4/rmi_driver.c
> index f17bfe0..27b8985 100644
> --- a/drivers/input/rmi4/rmi_driver.c
> +++ b/drivers/input/rmi4/rmi_driver.c
> @@ -21,6 +21,7 @@
> #include <linux/pm.h>
> #include <linux/slab.h>
> #include <linux/of.h>
> +#include <linux/firmware.h>
> #include <uapi/linux/input.h>
> #include <linux/rmi.h>
> #include "rmi_bus.h"
> @@ -41,7 +42,16 @@ static void rmi_free_function_list(struct rmi_device *rmi_dev)
>
> rmi_dbg(RMI_DEBUG_CORE, &rmi_dev->dev, "Freeing function list\n");
>
> + mutex_lock(&data->irq_mutex);
> +
> data->f01_container = NULL;
> + data->f34_container = NULL;
> + data->irq_status = NULL;
> + data->fn_irq_bits = NULL;
> + data->current_irq_mask = NULL;
> + data->new_irq_mask = NULL;
> +
> + mutex_unlock(&data->irq_mutex);
>
> /* Doing it in the reverse order so F01 will be removed last */
> list_for_each_entry_safe_reverse(fn, tmp,
> @@ -49,6 +59,7 @@ static void rmi_free_function_list(struct rmi_device *rmi_dev)
> list_del(&fn->node);
> rmi_unregister_function(fn);
> }
> +
> }
>
> static int reset_one_function(struct rmi_function *fn)
> @@ -143,8 +154,11 @@ int rmi_process_interrupt_requests(struct rmi_device *rmi_dev)
> struct rmi_function *entry;
> int error;
>
> - if (!data)
> + mutex_lock(&data->irq_mutex);
> + if (!data || !data->irq_status || !data->f01_container) {
> + mutex_unlock(&data->irq_mutex);

I have been now convinced that IRQ should be handled in core. It would
make everyone's life easier and I think means that we won't need these
checks given that IRQ could simply be disabled during FW update.

I guess it's time to resurrect Bjorn's patch at
https://lkml.org/lkml/2016/5/9/1055

> return 0;
> + }
>
> if (!rmi_dev->xport->attn_data) {
> error = rmi_read_block(rmi_dev,
> @@ -156,7 +170,6 @@ int rmi_process_interrupt_requests(struct rmi_device *rmi_dev)
> }
> }
>
> - mutex_lock(&data->irq_mutex);
> bitmap_and(data->irq_status, data->irq_status, data->current_irq_mask,
> data->irq_count);
> /*
> @@ -723,6 +736,7 @@ static int rmi_initial_reset(struct rmi_device *rmi_dev,
> return RMI_SCAN_DONE;
> }
>
> + rmi_dbg(RMI_DEBUG_CORE, &rmi_dev->dev, "Sending reset\n");
> error = rmi_write_block(rmi_dev, cmd_addr, &cmd_buf, 1);
> if (error) {
> dev_err(&rmi_dev->dev,
> @@ -779,6 +793,8 @@ static int rmi_create_function(struct rmi_device *rmi_dev,
>
> if (pdt->function_number == 0x01)
> data->f01_container = fn;
> + else if (pdt->function_number == 0x34)
> + data->f34_container = fn;

Not sure you need to have a special role for f34 in core (see below).

>
> list_add_tail(&fn->node, &data->function_list);
>
> @@ -815,10 +831,76 @@ int rmi_driver_resume(struct rmi_device *rmi_dev)
> }
> EXPORT_SYMBOL_GPL(rmi_driver_resume);
>
> +#ifdef CONFIG_RMI4_F34
> +static int rmi_firmware_update(struct rmi_driver_data *data,
> + const struct firmware *fw);
> +
> +static ssize_t rmi_driver_update_fw_store(struct device *dev,
> + struct device_attribute *dattr,
> + const char *buf, size_t count)
> +{
> + struct rmi_driver_data *data = dev_get_drvdata(dev);
> + char fw_name[NAME_MAX];
> + const struct firmware *fw;
> + size_t copy_count = count;
> + int ret;
> +
> + if (count == 0 || count >= NAME_MAX)
> + return -EINVAL;
> +
> + if (buf[count - 1] == '\0' || buf[count - 1] == '\n')
> + copy_count -= 1;
> +
> + strncpy(fw_name, buf, copy_count);
> + fw_name[copy_count] = '\0';
> +
> + ret = request_firmware(&fw, fw_name, dev);
> + if (ret)
> + return ret;
> +
> + ret = rmi_firmware_update(data, fw);
> +
> + release_firmware(fw);
> +
> + return ret ?: count;
> +}
> +
> +static DEVICE_ATTR(update_fw, S_IWUSR, NULL, rmi_driver_update_fw_store);
> +
> +static ssize_t rmi_driver_update_fw_status_show(struct device *dev,
> + struct device_attribute *dattr,
> + char *buf)
> +{
> + struct rmi_driver_data *data = dev_get_drvdata(dev);
> + int update_status = 0;
> +
> + if (data->f34_container)
> + update_status = rmi_f34_status(data->f34_container);
> +
> + return scnprintf(buf, PAGE_SIZE, "%d\n", update_status);
> +}
> +
> +static DEVICE_ATTR(update_fw_status, S_IRUGO,
> + rmi_driver_update_fw_status_show, NULL);
> +
> +static struct attribute *rmi_firmware_attrs[] = {
> + &dev_attr_update_fw.attr,
> + &dev_attr_update_fw_status.attr,
> + NULL
> +};
> +
> +static struct attribute_group rmi_firmware_attr_group = {
> + .attrs = rmi_firmware_attrs,
> +};

I don't like having those F34 specific sysfs path in rmi_driver.c. I
think the issue is that for rmi_firmware_update() you need to have
access to internals of rmi_driver, but I'd rather see those exported
someway so that the sysfs gets in rmi-f34.c

> +#endif /* CONFIG_RMI4_F34 */
> +
> static int rmi_driver_remove(struct device *dev)
> {
> struct rmi_device *rmi_dev = to_rmi_device(dev);
>
> +#ifdef CONFIG_RMI4_F34
> + sysfs_remove_group(&rmi_dev->dev.kobj, &rmi_firmware_attr_group);
> +#endif
> rmi_free_function_list(rmi_dev);
>
> return 0;
> @@ -923,6 +1005,75 @@ err_destroy_functions:
> return retval;
> }
>
> +#ifdef CONFIG_RMI4_F34
> +static int rmi_firmware_update(struct rmi_driver_data *data,
> + const struct firmware *fw)
> +{
> + struct device *dev = &data->rmi_dev->dev;
> + int ret;
> +
> + if (!data->f34_container) {
> + dev_warn(dev, "%s: No F34 present!\n",
> + __func__);
> + return -EINVAL;
> + }
> +
> + ret = rmi_f34_check_supported(data->f34_container);
> + if (ret)
> + return ret;
> +
> + /* Enter flash mode */
> + rmi_dbg(RMI_DEBUG_CORE, dev, "Enabling flash\n");
> + ret = rmi_f34_enable_flash(data->f34_container);
> + if (ret)
> + return ret;
> +
> + /* Tear down functions and re-probe */
> + rmi_free_function_list(data->rmi_dev);
> +
> + ret = rmi_probe_interrupts(data);
> + if (ret)
> + return ret;
> +
> + ret = rmi_init_functions(data);
> + if (ret)
> + return ret;
> +
> + if (!data->f01_bootloader_mode || !data->f34_container) {
> + dev_warn(dev, "%s: No F34 present or not in bootloader!\n",
> + __func__);
> + return -EINVAL;
> + }
> +
> + /* Perform firmware update */
> + ret = rmi_f34_update_firmware(data->f34_container, fw);
> +
> + /* Re-probe */
> + rmi_dbg(RMI_DEBUG_CORE, dev, "Re-probing device\n");
> + rmi_free_function_list(data->rmi_dev);
> +
> + ret = rmi_scan_pdt(data->rmi_dev, NULL, rmi_initial_reset);
> + if (ret < 0)
> + dev_warn(dev, "RMI reset failed!\n");
> +
> + ret = rmi_probe_interrupts(data);
> + if (ret)
> + return ret;
> +
> + ret = rmi_init_functions(data);
> + if (ret)
> + return ret;

You could basically export rmi_fn_reset() which would call
rmi_free_function_list(), rmi_scan_pdt (if initial reset),
rmi_probe_interrupts() and rmi_init_functions, and this would allow you
to have all this in f34.

Cheers,
Benjamin

> +
> + if (data->f01_container->dev.driver)
> + /* Driver already bound, so enable ATTN now. */
> + return enable_sensor(data->rmi_dev);
> +
> + rmi_dbg(RMI_DEBUG_CORE, dev, "%s complete\n", __func__);
> +
> + return ret;
> +}
> +#endif
> +
> static int rmi_driver_probe(struct device *dev)
> {
> struct rmi_driver *rmi_driver;
> @@ -1022,6 +1173,12 @@ static int rmi_driver_probe(struct device *dev)
> "%s/input0", dev_name(dev));
> }
>
> +#ifdef CONFIG_RMI4_F34
> + retval = sysfs_create_group(&dev->kobj, &rmi_firmware_attr_group);
> + if (retval)
> + goto err_destroy_functions;
> +#endif
> +
> retval = rmi_init_functions(data);
> if (retval)
> goto err;
> diff --git a/drivers/input/rmi4/rmi_driver.h b/drivers/input/rmi4/rmi_driver.h
> index 6e140fa..a9045e6 100644
> --- a/drivers/input/rmi4/rmi_driver.h
> +++ b/drivers/input/rmi4/rmi_driver.h
> @@ -96,10 +96,17 @@ bool rmi_is_physical_driver(struct device_driver *);
> int rmi_register_physical_driver(void);
> void rmi_unregister_physical_driver(void);
>
> +struct firmware;
> +
> char *rmi_f01_get_product_ID(struct rmi_function *fn);
> +int rmi_f34_update_firmware(struct rmi_function *fn, const struct firmware *fw);
> +int rmi_f34_enable_flash(struct rmi_function *fn);
> +int rmi_f34_status(struct rmi_function *fn);
> +int rmi_f34_check_supported(struct rmi_function *fn);
>
> extern struct rmi_function_handler rmi_f01_handler;
> extern struct rmi_function_handler rmi_f11_handler;
> extern struct rmi_function_handler rmi_f12_handler;
> extern struct rmi_function_handler rmi_f30_handler;
> +extern struct rmi_function_handler rmi_f34_handler;
> #endif
> diff --git a/drivers/input/rmi4/rmi_f01.c b/drivers/input/rmi4/rmi_f01.c
> index fac81fc..9bd2a9e 100644
> --- a/drivers/input/rmi4/rmi_f01.c
> +++ b/drivers/input/rmi4/rmi_f01.c
> @@ -63,6 +63,8 @@ struct f01_basic_properties {
> #define RMI_F01_STATUS_CODE(status) ((status) & 0x0f)
> /* The device has lost its configuration for some reason. */
> #define RMI_F01_STATUS_UNCONFIGURED(status) (!!((status) & 0x80))
> +/* The device is in bootloader mode */
> +#define RMI_F01_STATUS_BOOTLOADER(status) ((status) & 0x40)
>
> /* Control register bits */
>
> @@ -594,6 +596,10 @@ static int rmi_f01_attention(struct rmi_function *fn,
> return error;
> }
>
> + if (RMI_F01_STATUS_BOOTLOADER(device_status))
> + dev_warn(&fn->dev,
> + "Device in bootloader mode, please update firmware\n");
> +
> if (RMI_F01_STATUS_UNCONFIGURED(device_status)) {
> dev_warn(&fn->dev, "Device reset detected.\n");
> error = rmi_dev->driver->reset_handler(rmi_dev);
> diff --git a/drivers/input/rmi4/rmi_f34.c b/drivers/input/rmi4/rmi_f34.c
> new file mode 100644
> index 0000000..9a56695
> --- /dev/null
> +++ b/drivers/input/rmi4/rmi_f34.c
> @@ -0,0 +1,446 @@
> +/*
> + * Copyright (c) 2007-2016, Synaptics Incorporated
> + * Copyright (C) 2016 Zodiac Inflight Innovations
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms of the GNU General Public License version 2 as published by
> + * the Free Software Foundation.
> + */
> +
> +#include <linux/kernel.h>
> +#include <linux/rmi.h>
> +#include <linux/firmware.h>
> +#include <asm/unaligned.h>
> +#include <asm/unaligned.h>
> +
> +#include "rmi_driver.h"
> +
> +/* F34 image file offsets. */
> +#define F34_FW_IMAGE_OFFSET 0x100
> +
> +/* F34 register offsets. */
> +#define F34_BLOCK_DATA_OFFSET 2
> +
> +/* F34 commands */
> +#define F34_WRITE_FW_BLOCK 0x2
> +#define F34_ERASE_ALL 0x3
> +#define F34_READ_CONFIG_BLOCK 0x5
> +#define F34_WRITE_CONFIG_BLOCK 0x6
> +#define F34_ERASE_CONFIG 0x7
> +#define F34_ENABLE_FLASH_PROG 0xf
> +
> +#define F34_STATUS_IN_PROGRESS 0xff
> +#define F34_STATUS_IDLE 0x80
> +
> +#define F34_IDLE_WAIT_MS 500
> +#define F34_ENABLE_WAIT_MS 300
> +#define F34_ERASE_WAIT_MS 5000
> +
> +#define F34_BOOTLOADER_ID_LEN 2
> +
> +struct rmi_f34_firmware {
> + __le32 checksum;
> + u8 pad1[3];
> + u8 bootloader_version;
> + __le32 image_size;
> + __le32 config_size;
> + u8 product_id[10];
> + u8 product_info[2];
> + u8 pad2[228];
> + u8 data[];
> +};
> +
> +struct f34_data {
> + struct rmi_function *fn;
> +
> + u16 block_size;
> + u16 fw_blocks;
> + u16 config_blocks;
> + u16 ctrl_address;
> + u8 status;
> + struct completion cmd_done;
> +
> + struct mutex flash_mutex;
> +
> + int update_status;
> + int update_progress;
> + int update_size;
> + struct completion async_firmware_done;
> +
> + unsigned char bootloader_id[5];
> + unsigned char configuration_id[9];
> +};
> +
> +static int rmi_f34_write_bootloader_id(struct f34_data *f34)
> +{
> + struct rmi_function *fn = f34->fn;
> + struct rmi_device *rmi_dev = fn->rmi_dev;
> + u8 bootloader_id[F34_BOOTLOADER_ID_LEN];
> + int ret;
> +
> + ret = rmi_read_block(rmi_dev, fn->fd.query_base_addr,
> + bootloader_id, sizeof(bootloader_id));
> + if (ret) {
> + dev_err(&fn->dev, "%s: Reading bootloader ID failed: %d\n",
> + __func__, ret);
> + return ret;
> + }
> +
> + rmi_dbg(RMI_DEBUG_FN, &fn->dev, "%s: writing bootloader id '%c%c'\n",
> + __func__, bootloader_id[0], bootloader_id[1]);
> +
> + ret = rmi_write_block(rmi_dev,
> + fn->fd.data_base_addr + F34_BLOCK_DATA_OFFSET,
> + bootloader_id, sizeof(bootloader_id));
> + if (ret) {
> + dev_err(&fn->dev, "Failed to write bootloader ID: %d\n", ret);
> + return ret;
> + }
> +
> + return 0;
> +}
> +
> +static int rmi_f34_command(struct f34_data *f34, u8 command,
> + unsigned int timeout, bool write_bl_id)
> +{
> + struct rmi_function *fn = f34->fn;
> + struct rmi_device *rmi_dev = fn->rmi_dev;
> + int ret;
> +
> + if (write_bl_id) {
> + ret = rmi_f34_write_bootloader_id(f34);
> + if (ret)
> + return ret;
> + }
> +
> + init_completion(&f34->cmd_done);
> +
> + ret = rmi_read(rmi_dev, f34->ctrl_address, &f34->status);
> + if (ret) {
> + dev_err(&f34->fn->dev,
> + "%s: Failed to read cmd register: %d (command %#02x)\n",
> + __func__, ret, command);
> + return ret;
> + }
> +
> + f34->status |= command & 0x0f;
> +
> + ret = rmi_write(rmi_dev, f34->ctrl_address, f34->status);
> + if (ret < 0) {
> + dev_err(&f34->fn->dev,
> + "Failed to write F34 command %#02x: %d\n",
> + command, ret);
> + return ret;
> + }
> +
> + if (!wait_for_completion_timeout(&f34->cmd_done,
> + msecs_to_jiffies(timeout))) {
> +
> + ret = rmi_read(rmi_dev, f34->ctrl_address, &f34->status);
> + if (ret) {
> + dev_err(&f34->fn->dev,
> + "%s: failed to read status after command %#02x timed out: %d\n",
> + __func__, command, ret);
> + return ret;
> + }
> +
> + if (f34->status & 0x7f) {
> + dev_err(&f34->fn->dev,
> + "%s: command %#02x timed out, fw status: %#02x\n",
> + __func__, command, f34->status);
> + return -ETIMEDOUT;
> + }
> + }
> +
> + return 0;
> +}
> +
> +static int rmi_f34_attention(struct rmi_function *fn, unsigned long *irq_bits)
> +{
> + struct f34_data *f34 = dev_get_drvdata(&fn->dev);
> + int ret;
> +
> + ret = rmi_read(f34->fn->rmi_dev, f34->ctrl_address,
> + &f34->status);
> + rmi_dbg(RMI_DEBUG_FN, &fn->dev, "%s: status: %#02x, ret: %d\n",
> + __func__, f34->status, ret);
> +
> + if (!ret && !(f34->status & 0x7f))
> + complete(&f34->cmd_done);
> +
> + return 0;
> +}
> +
> +static int rmi_f34_write_blocks(struct f34_data *f34, const void *data,
> + int block_count, u8 command)
> +{
> + struct rmi_function *fn = f34->fn;
> + struct rmi_device *rmi_dev = fn->rmi_dev;
> + u16 address = fn->fd.data_base_addr + F34_BLOCK_DATA_OFFSET;
> + u8 start_address[] = { 0, 0 };
> + int i;
> + int ret;
> +
> + ret = rmi_write_block(rmi_dev, fn->fd.data_base_addr,
> + start_address, sizeof(start_address));
> + if (ret) {
> + dev_err(&fn->dev, "Failed to write initial zeros: %d\n", ret);
> + return ret;
> + }
> +
> + for (i = 0; i < block_count; i++) {
> + ret = rmi_write_block(rmi_dev, address, data, f34->block_size);
> + if (ret) {
> + dev_err(&fn->dev,
> + "failed to write block #%d: %d\n", i, ret);
> + return ret;
> + }
> +
> + ret = rmi_f34_command(f34, command, F34_IDLE_WAIT_MS, false);
> + if (ret) {
> + dev_err(&fn->dev,
> + "Failed to write command for block #%d: %d\n",
> + i, ret);
> + return ret;
> + }
> +
> + rmi_dbg(RMI_DEBUG_FN, &fn->dev, "wrote block %d of %d\n",
> + i + 1, block_count);
> +
> + data += f34->block_size;
> + f34->update_progress += f34->block_size;
> + f34->update_status = (f34->update_progress * 100) /
> + f34->update_size;
> + }
> +
> + return 0;
> +}
> +
> +static int rmi_f34_write_firmware(struct f34_data *f34, const void *data)
> +{
> + return rmi_f34_write_blocks(f34, data, f34->fw_blocks,
> + F34_WRITE_FW_BLOCK);
> +}
> +
> +static int rmi_f34_write_config(struct f34_data *f34, const void *data)
> +{
> + return rmi_f34_write_blocks(f34, data, f34->config_blocks,
> + F34_WRITE_CONFIG_BLOCK);
> +}
> +
> +int rmi_f34_enable_flash(struct rmi_function *fn)
> +{
> + struct f34_data *f34 = dev_get_drvdata(&fn->dev);
> +
> + return rmi_f34_command(f34, F34_ENABLE_FLASH_PROG,
> + F34_ENABLE_WAIT_MS, true);
> +}
> +
> +static int rmi_f34_flash_firmware(struct f34_data *f34,
> + const struct rmi_f34_firmware *syn_fw)
> +{
> + struct rmi_function *fn = f34->fn;
> + int ret;
> +
> + f34->update_progress = 0;
> + f34->update_size = syn_fw->image_size + syn_fw->config_size;
> + if (syn_fw->image_size) {
> + dev_info(&fn->dev, "Erasing FW...\n");
> + ret = rmi_f34_command(f34, F34_ERASE_ALL,
> + F34_ERASE_WAIT_MS, true);
> + if (ret)
> + return ret;
> +
> + dev_info(&fn->dev, "Writing firmware data (%d bytes)...\n",
> + syn_fw->image_size);
> + ret = rmi_f34_write_firmware(f34, syn_fw->data);
> + if (ret)
> + return ret;
> + }
> +
> + if (syn_fw->config_size) {
> + /*
> + * We only need to erase config if we haven't updated
> + * firmware.
> + */
> + if (!syn_fw->image_size) {
> + dev_info(&fn->dev, "%s: Erasing config data...\n",
> + __func__);
> + ret = rmi_f34_command(f34, F34_ERASE_CONFIG,
> + F34_ERASE_WAIT_MS, true);
> + if (ret)
> + return ret;
> + }
> +
> + dev_info(&fn->dev, "%s: Writing config data (%d bytes)...\n",
> + __func__, syn_fw->config_size);
> + ret = rmi_f34_write_config(f34,
> + &syn_fw->data[syn_fw->image_size]);
> + if (ret)
> + return ret;
> + }
> +
> + dev_info(&fn->dev, "%s: Firmware update complete\n", __func__);
> + return 0;
> +}
> +
> +int rmi_f34_update_firmware(struct rmi_function *fn, const struct firmware *fw)
> +{
> + struct f34_data *f34 = dev_get_drvdata(&fn->dev);
> + const struct rmi_f34_firmware *syn_fw;
> + int ret;
> +
> + syn_fw = (const struct rmi_f34_firmware *)fw->data;
> + BUILD_BUG_ON(offsetof(struct rmi_f34_firmware, data) !=
> + F34_FW_IMAGE_OFFSET);
> +
> + rmi_dbg(RMI_DEBUG_FN, &fn->dev,
> + "FW size:%d, checksum:%08x, image_size:%d, config_size:%d\n",
> + (int)fw->size,
> + le32_to_cpu(syn_fw->checksum),
> + le32_to_cpu(syn_fw->image_size),
> + le32_to_cpu(syn_fw->config_size));
> +
> + dev_info(&fn->dev,
> + "FW bootloader_id:%02x, product_id:%.*s, info: %02x%02x\n",
> + syn_fw->bootloader_version,
> + (int)sizeof(syn_fw->product_id), syn_fw->product_id,
> + syn_fw->product_info[0], syn_fw->product_info[1]);
> +
> + if (syn_fw->image_size &&
> + syn_fw->image_size != f34->fw_blocks * f34->block_size) {
> + dev_err(&fn->dev,
> + "Bad firmware image: fw size %d, expected %d\n",
> + syn_fw->image_size,
> + f34->fw_blocks * f34->block_size);
> + ret = -EILSEQ;
> + goto out;
> + }
> +
> + if (syn_fw->config_size &&
> + syn_fw->config_size != f34->config_blocks * f34->block_size) {
> + dev_err(&fn->dev,
> + "Bad firmware image: config size %d, expected %d\n",
> + syn_fw->config_size,
> + f34->config_blocks * f34->block_size);
> + ret = -EILSEQ;
> + goto out;
> + }
> +
> + if (syn_fw->image_size && !syn_fw->config_size) {
> + dev_err(&fn->dev,
> + "Bad firmware image: no config data\n");
> + ret = -EILSEQ;
> + goto out;
> + }
> +
> + dev_info(&fn->dev, "Starting firmware update\n");
> + mutex_lock(&f34->flash_mutex);
> +
> + ret = rmi_f34_flash_firmware(f34, syn_fw);
> + dev_info(&fn->dev, "Firmware update complete, status:%d\n", ret);
> +
> + f34->update_status = ret;
> + mutex_unlock(&f34->flash_mutex);
> +
> +out:
> + return ret;
> +}
> +
> +int rmi_f34_status(struct rmi_function *fn)
> +{
> + struct f34_data *f34 = dev_get_drvdata(&fn->dev);
> +
> + /*
> + * The status is the percentage complete, or once complete,
> + * zero for success or a negative return code.
> + */
> + return f34->update_status;
> +}
> +
> +int rmi_f34_check_supported(struct rmi_function *fn)
> +{
> + u8 version = fn->fd.function_version;
> +
> + /* Only version 0 currently supported */
> + if (version == 0) {
> + return 0;
> + } else {
> + dev_warn(&fn->dev, "F34 V%d not supported!\n", version);
> + return -ENODEV;
> + }
> +}
> +
> +static int rmi_f34_probe(struct rmi_function *fn)
> +{
> + struct f34_data *f34;
> + unsigned char f34_queries[9];
> + bool has_config_id;
> + int ret;
> +
> + ret = rmi_f34_check_supported(fn);
> + if (ret)
> + return ret;
> +
> + f34 = devm_kzalloc(&fn->dev, sizeof(struct f34_data), GFP_KERNEL);
> + if (!f34)
> + return -ENOMEM;
> +
> + f34->fn = fn;
> + dev_set_drvdata(&fn->dev, f34);
> +
> + mutex_init(&f34->flash_mutex);
> + init_completion(&f34->cmd_done);
> + init_completion(&f34->async_firmware_done);
> +
> + ret = rmi_read_block(fn->rmi_dev, fn->fd.query_base_addr,
> + f34_queries, sizeof(f34_queries));
> + if (ret) {
> + dev_err(&fn->dev, "%s: Failed to query properties\n",
> + __func__);
> + return ret;
> + }
> +
> + snprintf(f34->bootloader_id, sizeof(f34->bootloader_id),
> + "%c%c", f34_queries[0], f34_queries[1]);
> +
> + f34->block_size = get_unaligned_le16(&f34_queries[3]);
> + f34->fw_blocks = get_unaligned_le16(&f34_queries[5]);
> + f34->config_blocks = get_unaligned_le16(&f34_queries[7]);
> + f34->ctrl_address = fn->fd.data_base_addr + F34_BLOCK_DATA_OFFSET +
> + f34->block_size;
> + has_config_id = f34_queries[2] & (1 << 2);
> +
> + dev_info(&fn->dev, "Bootloader ID: %s\n", f34->bootloader_id);
> + rmi_dbg(RMI_DEBUG_FN, &fn->dev, "Block size: %d\n", f34->block_size);
> + rmi_dbg(RMI_DEBUG_FN, &fn->dev, "FW blocks: %d\n", f34->fw_blocks);
> + rmi_dbg(RMI_DEBUG_FN, &fn->dev, "CFG blocks: %d\n", f34->config_blocks);
> +
> + if (has_config_id) {
> + ret = rmi_read_block(fn->rmi_dev, fn->fd.control_base_addr,
> + f34_queries, sizeof(f34_queries));
> + if (ret) {
> + dev_err(&fn->dev, "Failed to read F34 config ID\n");
> + return ret;
> + }
> +
> + snprintf(f34->configuration_id,
> + sizeof(f34->configuration_id),
> + "%02x%02x%02x%02x", f34_queries[0],
> + f34_queries[1], f34_queries[2],
> + f34_queries[3]);
> + dev_info(&fn->dev, "Configuration ID: %s\n",
> + f34->configuration_id);
> + }
> +
> + return 0;
> +}
> +
> +struct rmi_function_handler rmi_f34_handler = {
> + .driver = {
> + .name = "rmi4_f34",
> + },
> + .func = 0x34,
> + .probe = rmi_f34_probe,
> + .attention = rmi_f34_attention,
> +};
> diff --git a/include/linux/rmi.h b/include/linux/rmi.h
> index e0aca14..a283a67 100644
> --- a/include/linux/rmi.h
> +++ b/include/linux/rmi.h
> @@ -330,6 +330,7 @@ struct rmi_driver_data {
> struct rmi_device *rmi_dev;
>
> struct rmi_function *f01_container;
> + struct rmi_function *f34_container;
> bool f01_bootloader_mode;
>
> u32 attn_count;
> --
> 2.7.4
>


2016-10-12 18:57:36

by Nick Dyer

[permalink] [raw]
Subject: Re: [PATCH v3 2/2] Input: synaptics-rmi4 - add support for F34 device reflash

Hi Benjamin-

Thanks for the review, I've answered in line.

On Mon, Oct 10, 2016 at 03:48:07PM +0200, Benjamin Tissoires wrote:
> On Sep 20 2016 or thereabouts, Nick Dyer wrote:
> > Signed-off-by: Nick Dyer <[email protected]>
> > Tested-by: Chris Healy <[email protected]>
> > ---
> > drivers/input/rmi4/Kconfig | 11 +
> > drivers/input/rmi4/Makefile | 1 +
> > drivers/input/rmi4/rmi_bus.c | 3 +
> > drivers/input/rmi4/rmi_driver.c | 161 ++++++++++++++-
> > drivers/input/rmi4/rmi_driver.h | 7 +
> > drivers/input/rmi4/rmi_f01.c | 6 +
> > drivers/input/rmi4/rmi_f34.c | 446 ++++++++++++++++++++++++++++++++++++++++
> > include/linux/rmi.h | 1 +
> > 8 files changed, 634 insertions(+), 2 deletions(-)
> > create mode 100644 drivers/input/rmi4/rmi_f34.c
[...]
> > @@ -143,8 +154,11 @@ int rmi_process_interrupt_requests(struct rmi_device *rmi_dev)
> > struct rmi_function *entry;
> > int error;
> >
> > - if (!data)
> > + mutex_lock(&data->irq_mutex);
> > + if (!data || !data->irq_status || !data->f01_container) {
> > + mutex_unlock(&data->irq_mutex);
>
> I have been now convinced that IRQ should be handled in core. It would
> make everyone's life easier and I think means that we won't need these
> checks given that IRQ could simply be disabled during FW update.
>
> I guess it's time to resurrect Bjorn's patch at
> https://lkml.org/lkml/2016/5/9/1055

We do use the interrupts in FW update in the current version. I haven't
done the measurements, but my expectation would be that performance is
slightly improved over polling, and it's more elegant.

> > +#ifdef CONFIG_RMI4_F34
> > +static int rmi_firmware_update(struct rmi_driver_data *data,
> > + const struct firmware *fw)
> > +{
> > + struct device *dev = &data->rmi_dev->dev;
> > + int ret;
> > +
> > + if (!data->f34_container) {
> > + dev_warn(dev, "%s: No F34 present!\n",
> > + __func__);
> > + return -EINVAL;
> > + }
> > +
> > + ret = rmi_f34_check_supported(data->f34_container);
> > + if (ret)
> > + return ret;
> > +
> > + /* Enter flash mode */
> > + rmi_dbg(RMI_DEBUG_CORE, dev, "Enabling flash\n");
> > + ret = rmi_f34_enable_flash(data->f34_container);
> > + if (ret)
> > + return ret;
> > +
> > + /* Tear down functions and re-probe */
> > + rmi_free_function_list(data->rmi_dev);
> > +
> > + ret = rmi_probe_interrupts(data);
> > + if (ret)
> > + return ret;
> > +
> > + ret = rmi_init_functions(data);
> > + if (ret)
> > + return ret;
> > +
> > + if (!data->f01_bootloader_mode || !data->f34_container) {
> > + dev_warn(dev, "%s: No F34 present or not in bootloader!\n",
> > + __func__);
> > + return -EINVAL;
> > + }
> > +
> > + /* Perform firmware update */
> > + ret = rmi_f34_update_firmware(data->f34_container, fw);
> > +
> > + /* Re-probe */
> > + rmi_dbg(RMI_DEBUG_CORE, dev, "Re-probing device\n");
> > + rmi_free_function_list(data->rmi_dev);
> > +
> > + ret = rmi_scan_pdt(data->rmi_dev, NULL, rmi_initial_reset);
> > + if (ret < 0)
> > + dev_warn(dev, "RMI reset failed!\n");
> > +
> > + ret = rmi_probe_interrupts(data);
> > + if (ret)
> > + return ret;
> > +
> > + ret = rmi_init_functions(data);
> > + if (ret)
> > + return ret;
>
> You could basically export rmi_fn_reset() which would call
> rmi_free_function_list(), rmi_scan_pdt (if initial reset),
> rmi_probe_interrupts() and rmi_init_functions, and this would allow you
> to have all this in f34.

I see what you mean, and I do agree that it would be neater to have all
of this in the f34 code.

However, the problem is that when you call rmi_free_function_list(), the
f34 driver and all the context information attached to it (struct
rmi_function, struct f34_data, and any sysfs attributes in the f34
directory) gets torn down, so you're kind of left without the branch you
were sitting on.

To get around that, I'd have to make f34 a special case anyway. Taking
that into account, the current solution seemed neater to me. I could
possibly cram a little more of it into rmi_f34.c, but I think the
context has to be "struct rmi_driver_data".

Nick

2016-10-13 13:53:37

by Benjamin Tissoires

[permalink] [raw]
Subject: Re: [PATCH v3 2/2] Input: synaptics-rmi4 - add support for F34 device reflash

On Oct 12 2016 or thereabouts, Nick Dyer wrote:
> Hi Benjamin-
>
> Thanks for the review, I've answered in line.
>
> On Mon, Oct 10, 2016 at 03:48:07PM +0200, Benjamin Tissoires wrote:
> > On Sep 20 2016 or thereabouts, Nick Dyer wrote:
> > > Signed-off-by: Nick Dyer <[email protected]>
> > > Tested-by: Chris Healy <[email protected]>
> > > ---
> > > drivers/input/rmi4/Kconfig | 11 +
> > > drivers/input/rmi4/Makefile | 1 +
> > > drivers/input/rmi4/rmi_bus.c | 3 +
> > > drivers/input/rmi4/rmi_driver.c | 161 ++++++++++++++-
> > > drivers/input/rmi4/rmi_driver.h | 7 +
> > > drivers/input/rmi4/rmi_f01.c | 6 +
> > > drivers/input/rmi4/rmi_f34.c | 446 ++++++++++++++++++++++++++++++++++++++++
> > > include/linux/rmi.h | 1 +
> > > 8 files changed, 634 insertions(+), 2 deletions(-)
> > > create mode 100644 drivers/input/rmi4/rmi_f34.c
> [...]
> > > @@ -143,8 +154,11 @@ int rmi_process_interrupt_requests(struct rmi_device *rmi_dev)
> > > struct rmi_function *entry;
> > > int error;
> > >
> > > - if (!data)
> > > + mutex_lock(&data->irq_mutex);
> > > + if (!data || !data->irq_status || !data->f01_container) {
> > > + mutex_unlock(&data->irq_mutex);
> >
> > I have been now convinced that IRQ should be handled in core. It would
> > make everyone's life easier and I think means that we won't need these
> > checks given that IRQ could simply be disabled during FW update.
> >
> > I guess it's time to resurrect Bjorn's patch at
> > https://lkml.org/lkml/2016/5/9/1055
>
> We do use the interrupts in FW update in the current version. I haven't
> done the measurements, but my expectation would be that performance is
> slightly improved over polling, and it's more elegant.
>
> > > +#ifdef CONFIG_RMI4_F34
> > > +static int rmi_firmware_update(struct rmi_driver_data *data,
> > > + const struct firmware *fw)
> > > +{
> > > + struct device *dev = &data->rmi_dev->dev;
> > > + int ret;
> > > +
> > > + if (!data->f34_container) {
> > > + dev_warn(dev, "%s: No F34 present!\n",
> > > + __func__);
> > > + return -EINVAL;
> > > + }
> > > +
> > > + ret = rmi_f34_check_supported(data->f34_container);
> > > + if (ret)
> > > + return ret;
> > > +
> > > + /* Enter flash mode */
> > > + rmi_dbg(RMI_DEBUG_CORE, dev, "Enabling flash\n");
> > > + ret = rmi_f34_enable_flash(data->f34_container);
> > > + if (ret)
> > > + return ret;
> > > +
> > > + /* Tear down functions and re-probe */
> > > + rmi_free_function_list(data->rmi_dev);
> > > +
> > > + ret = rmi_probe_interrupts(data);
> > > + if (ret)
> > > + return ret;
> > > +
> > > + ret = rmi_init_functions(data);
> > > + if (ret)
> > > + return ret;
> > > +
> > > + if (!data->f01_bootloader_mode || !data->f34_container) {
> > > + dev_warn(dev, "%s: No F34 present or not in bootloader!\n",
> > > + __func__);
> > > + return -EINVAL;
> > > + }
> > > +
> > > + /* Perform firmware update */
> > > + ret = rmi_f34_update_firmware(data->f34_container, fw);
> > > +
> > > + /* Re-probe */
> > > + rmi_dbg(RMI_DEBUG_CORE, dev, "Re-probing device\n");
> > > + rmi_free_function_list(data->rmi_dev);
> > > +
> > > + ret = rmi_scan_pdt(data->rmi_dev, NULL, rmi_initial_reset);
> > > + if (ret < 0)
> > > + dev_warn(dev, "RMI reset failed!\n");
> > > +
> > > + ret = rmi_probe_interrupts(data);
> > > + if (ret)
> > > + return ret;
> > > +
> > > + ret = rmi_init_functions(data);
> > > + if (ret)
> > > + return ret;
> >
> > You could basically export rmi_fn_reset() which would call
> > rmi_free_function_list(), rmi_scan_pdt (if initial reset),
> > rmi_probe_interrupts() and rmi_init_functions, and this would allow you
> > to have all this in f34.
>
> I see what you mean, and I do agree that it would be neater to have all
> of this in the f34 code.
>
> However, the problem is that when you call rmi_free_function_list(), the
> f34 driver and all the context information attached to it (struct
> rmi_function, struct f34_data, and any sysfs attributes in the f34
> directory) gets torn down, so you're kind of left without the branch you
> were sitting on.
>
> To get around that, I'd have to make f34 a special case anyway. Taking
> that into account, the current solution seemed neater to me. I could
> possibly cram a little more of it into rmi_f34.c, but I think the
> context has to be "struct rmi_driver_data".

If I understand correctly, rmi_firmware_update() is only called through
the sysfs. So how about you export the required functions from core you
are using and export 2 functions from rmi_f34 that will be a special
case: rmi_f34_create_sysfs() and rmi_f34_remove_sysfs() (or any better
names). You could just put your code in rmi_f34, provide noops
declarations if RMI_F34 is not set, and core will have only 2 calls to
rmi_f34.

BTW, I am thinking at carrying in my next RMI4 series your 1/2 patch. I
also want to take Bjorn and Andrew left patches so that we have a common
tree at some point. Any objections?
Of course, if you resubmit before me, feel free to carry over 1/2.

Cheers,
Benjamin

>
> Nick

2016-10-13 20:36:46

by Nick Dyer

[permalink] [raw]
Subject: Re: [PATCH v3 2/2] Input: synaptics-rmi4 - add support for F34 device reflash

On Thu, Oct 13, 2016 at 03:52:45PM +0200, Benjamin Tissoires wrote:
> > > You could basically export rmi_fn_reset() which would call
> > > rmi_free_function_list(), rmi_scan_pdt (if initial reset),
> > > rmi_probe_interrupts() and rmi_init_functions, and this would allow you
> > > to have all this in f34.
> >
> > I see what you mean, and I do agree that it would be neater to have all
> > of this in the f34 code.
> >
> > However, the problem is that when you call rmi_free_function_list(), the
> > f34 driver and all the context information attached to it (struct
> > rmi_function, struct f34_data, and any sysfs attributes in the f34
> > directory) gets torn down, so you're kind of left without the branch you
> > were sitting on.
> >
> > To get around that, I'd have to make f34 a special case anyway. Taking
> > that into account, the current solution seemed neater to me. I could
> > possibly cram a little more of it into rmi_f34.c, but I think the
> > context has to be "struct rmi_driver_data".
>
> If I understand correctly, rmi_firmware_update() is only called through
> the sysfs. So how about you export the required functions from core you
> are using and export 2 functions from rmi_f34 that will be a special
> case: rmi_f34_create_sysfs() and rmi_f34_remove_sysfs() (or any better
> names). You could just put your code in rmi_f34, provide noops
> declarations if RMI_F34 is not set, and core will have only 2 calls to
> rmi_f34.

OK: I will have a go at achieving this over the next few days. I also
have some changes almost ready to add F34 bootloader V7 support.

> BTW, I am thinking at carrying in my next RMI4 series your 1/2 patch. I
> also want to take Bjorn and Andrew left patches so that we have a common
> tree at some point. Any objections?
> Of course, if you resubmit before me, feel free to carry over 1/2.

Sounds like a good idea, it will reduce merging difficulty later.

N