Re: [PATCH v6 2/2] efi: a misc char interface for user to update efi firmware

From: Borislav Petkov
Date: Fri Oct 02 2015 - 13:39:19 EST


On Fri, Oct 02, 2015 at 05:05:54AM +0800, Kweh, Hock Leong wrote:
> From: "Kweh, Hock Leong" <hock.leong.kweh@xxxxxxxxx>
>
> Introducing a kernel module to expose capsule loader interface
> (misc char device file note) for user to upload capsule binaries.
>
> Example method to load the capsule binary:
> cat firmware.bin > /dev/efi_capsule_loader
>
> Cc: Matt Fleming <matt.fleming@xxxxxxxxx>
> Signed-off-by: Kweh, Hock Leong <hock.leong.kweh@xxxxxxxxx>
> ---
> drivers/firmware/efi/Kconfig | 10 ++
> drivers/firmware/efi/Makefile | 1 +
> drivers/firmware/efi/efi-capsule-loader.c | 246 +++++++++++++++++++++++++++++
> 3 files changed, 257 insertions(+)
> create mode 100644 drivers/firmware/efi/efi-capsule-loader.c
>
> diff --git a/drivers/firmware/efi/Kconfig b/drivers/firmware/efi/Kconfig
> index f712d47..0be8ee3 100644
> --- a/drivers/firmware/efi/Kconfig
> +++ b/drivers/firmware/efi/Kconfig
> @@ -60,6 +60,16 @@ config EFI_RUNTIME_WRAPPERS
> config EFI_ARMSTUB
> bool
>
> +config EFI_CAPSULE_LOADER
> + tristate "EFI capsule loader"
> + depends on EFI
> + help
> + This option exposes a loader interface "/dev/efi_capsule_loader" for
> + user to load EFI capsule binary and update the EFI firmware through
> + system reboot.

Do you mean here: user has to cat the fw binary *and* *then* reboot the box?

If so, we probably should issue such note to dmesg after successful
write...

> +
> + If unsure, say N.
> +
> endmenu
>
> config UEFI_CPER
> diff --git a/drivers/firmware/efi/Makefile b/drivers/firmware/efi/Makefile
> index 698846e..5ab031a 100644
> --- a/drivers/firmware/efi/Makefile
> +++ b/drivers/firmware/efi/Makefile
> @@ -8,3 +8,4 @@ obj-$(CONFIG_UEFI_CPER) += cper.o
> obj-$(CONFIG_EFI_RUNTIME_MAP) += runtime-map.o
> obj-$(CONFIG_EFI_RUNTIME_WRAPPERS) += runtime-wrappers.o
> obj-$(CONFIG_EFI_STUB) += libstub/
> +obj-$(CONFIG_EFI_CAPSULE_LOADER) += efi-capsule-loader.o
> diff --git a/drivers/firmware/efi/efi-capsule-loader.c b/drivers/firmware/efi/efi-capsule-loader.c
> new file mode 100644
> index 0000000..571e0c8
> --- /dev/null
> +++ b/drivers/firmware/efi/efi-capsule-loader.c
> @@ -0,0 +1,246 @@
> +/*
> + * EFI capsule loader driver.
> + *
> + * Copyright 2015 Intel Corporation
> + *
> + * This file is part of the Linux kernel, and is made available under
> + * the terms of the GNU General Public License version 2.
> + */
> +
> +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> +
> +#include <linux/kernel.h>
> +#include <linux/module.h>
> +#include <linux/miscdevice.h>
> +#include <linux/highmem.h>
> +#include <linux/slab.h>
> +#include <linux/mutex.h>
> +#include <linux/efi.h>
> +
> +#define DEV_NAME "efi_capsule_loader"
> +
> +static struct capsule_info {
> + char header_obtained;
> + long index;
> + unsigned long count;
> + unsigned long total_size;
> + struct page **pages;
> + unsigned long page_count_remain;
> +} cap_info;
> +
> +static DEFINE_MUTEX(capsule_loader_lock);
> +
> +/*
> + * This function will free the current & all previous allocated buffer pages.

No need to start the comment with "This function" - that's implicitly the case.
Same for the other comments below.

> + * The input parameter is the allocated buffer page for current execution which
> + * has not yet assigned to pages array. The input param can be NULL if the
> + * current execution has not allocated any buffer page.

We have the standard format of documenting those:

* @kbuff_page: the allocated buffer page...

> + */
> +static void efi_free_all_buff_pages(struct page *kbuff_page)
> +{
> + if (!kbuff_page)
> + __free_page(kbuff_page);
> +
> + if (cap_info.index > 0)
> + while (cap_info.index > 0)
> + __free_page(cap_info.pages[--cap_info.index]);
> +
> + if (cap_info.header_obtained)
> + kfree(cap_info.pages);
> +
> + /* indicate to the next that error had occurred */
> + cap_info.index = -2;
> +}
> +
> +/*
> + * This function will store the capsule binary and pass it to
> + * efi_capsule_update() API in capsule.c.
> + *
> + * Expectation:
> + * - userspace tool should start at the beginning of capsule binary and pass
> + * data in sequential.
> + * - user should close and re-open this file note in order to upload more
> + * capsules.
> + * - any error occurred, user should close the file and restart the operation
> + * for the next try otherwise -EIO will be returned until the file is closed.
> + */
> +static ssize_t efi_capsule_write(struct file *file, const char __user *buff,
> + size_t count, loff_t *offp)
> +{
> + int ret = 0;
> + struct page *kbuff_page;
> + void *kbuff;
> + size_t write_byte;
> +
> + pr_debug("%s: Entering Write()\n", __func__);
> + if (count == 0)
> + return 0;
> +
> + /* return error while error had occurred or capsule uploading is done */
> + if (cap_info.index < 0)
> + return -EIO;
> +
> + /* only alloc a new page when the current page is full */
> + if (!cap_info.page_count_remain) {
> + kbuff_page = alloc_page(GFP_KERNEL);
> + if (!kbuff_page) {
> + pr_debug("%s: alloc_page() failed\n", __func__);
> + efi_free_all_buff_pages(NULL);
> + return -ENOMEM;
> + }
> + cap_info.page_count_remain = PAGE_SIZE;
> + } else {
> + kbuff_page = cap_info.pages[--cap_info.index];
> + }
> + kbuff = kmap(kbuff_page);
> + if (!kbuff) {
> + pr_debug("%s: kmap() failed\n", __func__);
> + efi_free_all_buff_pages(kbuff_page);
> + return -EFAULT;
> + }
> + kbuff += PAGE_SIZE - cap_info.page_count_remain;
> +
> + /* copy capsule binary data from user space to kernel space buffer */
> + write_byte = min_t(size_t, count, cap_info.page_count_remain);
> + if (copy_from_user(kbuff, buff, write_byte)) {
> + pr_debug("%s: copy_from_user() failed\n", __func__);
> + kunmap(kbuff_page);
> + efi_free_all_buff_pages(kbuff_page);
> + return -EFAULT;
> + }
> + cap_info.page_count_remain -= write_byte;
> +
> + /* setup capsule binary info structure */
> + if (cap_info.header_obtained == 0 && cap_info.index == 0) {
> + efi_capsule_header_t *cap_hdr = kbuff;
> + int reset_type;
> + size_t pages_needed = ALIGN(cap_hdr->imagesize, PAGE_SIZE) >>
> + PAGE_SHIFT;
> +
> + if (pages_needed == 0) {
> + pr_err("%s: pages count invalid\n", __func__);
> + kunmap(kbuff_page);
> + efi_free_all_buff_pages(kbuff_page);
> + return -EINVAL;
> + }
> +
> + /* check if the capsule binary supported */
> + ret = efi_capsule_supported(cap_hdr->guid, cap_hdr->flags,
> + cap_hdr->imagesize, &reset_type);
> + if (ret) {
> + pr_err("%s: efi_capsule_supported() failed\n",
> + __func__);
> + kunmap(kbuff_page);
> + efi_free_all_buff_pages(kbuff_page);
> + return ret;
> + }
> +
> + cap_info.total_size = cap_hdr->imagesize;
> + cap_info.pages = kmalloc_array(pages_needed, sizeof(void *),
> + GFP_KERNEL);
> + if (!cap_info.pages) {
> + pr_debug("%s: kmalloc_array() failed\n", __func__);
> + kunmap(kbuff_page);
> + efi_free_all_buff_pages(kbuff_page);
> + return -ENOMEM;
> + }
> +
> + cap_info.header_obtained = 1;
> + }
> +
> + cap_info.pages[cap_info.index++] = kbuff_page;
> + cap_info.count += write_byte;
> + kunmap(kbuff_page);
> +
> + /* submit the full binary to efi_capsule_update() API */
> + if (cap_info.count >= cap_info.total_size) {
> + void *cap_hdr_temp;
> +
> + cap_hdr_temp = kmap(cap_info.pages[0]);
> + if (!cap_hdr_temp) {
> + pr_debug("%s: kmap() failed\n", __func__);
> + efi_free_all_buff_pages(NULL);
> + return -EFAULT;
> + }
> + ret = efi_capsule_update(cap_hdr_temp, cap_info.pages);
> + kunmap(cap_info.pages[0]);
> + if (ret) {
> + pr_err("%s: efi_capsule_update() failed\n", __func__);
> + efi_free_all_buff_pages(NULL);
> + return ret;
> + }
> + /* indicate capsule binary uploading is done */
> + cap_info.index = -1;
> + }
> +
> + return write_byte;
> +}

This function is huuuge and it could use some splitting. At least the
code blocks under

/* setup capsule binary info structure */

and

/* submit the full binary to efi_capsule_update() API */

could be separate, helper functions.

And you have the same repeated pattern on the error path:

kunmap();
efi_free_all_buff_pages(NULL);
return -<errno>

Mind rewriting that with goto err_... labels like it is normally done in
the kernel.

> +
> +static int efi_capsule_release(struct inode *inode, struct file *file)
> +{
> + int ret = 0;
> +
> + pr_debug("%s: Exit in Release()\n", __func__);
> + /* release uncompleted capsule upload */

That should be above the function.

> + if (cap_info.index > 0) {
> + pr_err("%s: capsule upload not complete\n", __func__);
> + efi_free_all_buff_pages(NULL);
> + ret = -ECANCELED;
> + }
> + mutex_unlock(&capsule_loader_lock);
> +
> + /*
> + * we would not free the successful submitted buffer pages here due to
> + * efi update require system reboot with data maintained.
> + */

That too.

> + return ret;
> +}
> +
> +static int efi_capsule_open(struct inode *inode, struct file *file)
> +{
> + pr_debug("%s: Entering Open()\n", __func__);

Do we really need those?

> + if (!mutex_trylock(&capsule_loader_lock))
> + return -EBUSY;
> +
> + memset(&cap_info, 0, sizeof(cap_info));
> + return 0;
> +}
> +
> +static const struct file_operations efi_capsule_fops = {
> + .owner = THIS_MODULE,
> + .open = efi_capsule_open,
> + .write = efi_capsule_write,
> + .release = efi_capsule_release,
> + .llseek = no_llseek,
> +};
> +
> +static struct miscdevice efi_capsule_misc = {
> + .minor = MISC_DYNAMIC_MINOR,
> + .name = DEV_NAME,
> + .fops = &efi_capsule_fops,
> +};
> +
> +static int __init efi_capsule_loader_init(void)
> +{
> + int ret;
> +
> + mutex_init(&capsule_loader_lock);
> +
> + ret = misc_register(&efi_capsule_misc);
> + if (ret)
> + pr_err("%s: Failed to register misc char file note\n",
> + __func__);

mutex_destroy() on the error path.

> +
> + return ret;
> +}
> +module_init(efi_capsule_loader_init);
> +
> +static void __exit efi_capsule_loader_exit(void)
> +{
> + mutex_destroy(&capsule_loader_lock);
> + misc_deregister(&efi_capsule_misc);

Those are in the wrong order - you need to destroy the mutex last.

> +}
> +module_exit(efi_capsule_loader_exit);
> +
> +MODULE_DESCRIPTION("EFI capsule firmware binary loader");
> +MODULE_LICENSE("GPL v2");
> --
> 1.7.9.5
>
>

--
Regards/Gruss,
Boris.

ECO tip #101: Trim your mails when you reply.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/