dect
/
linux-2.6
Archived
13
0
Fork 0
This repository has been archived on 2022-02-17. You can view files and clone it, but cannot push or open issues or pull requests.
linux-2.6/drivers/firmware/efivars.c

1835 lines
45 KiB
C
Raw Permalink Normal View History

/*
* EFI Variables - efivars.c
*
* Copyright (C) 2001,2003,2004 Dell <Matt_Domsch@dell.com>
* Copyright (C) 2004 Intel Corporation <matthew.e.tolentino@intel.com>
*
* This code takes all variables accessible from EFI runtime and
* exports them via sysfs
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
* Changelog:
*
* 17 May 2004 - Matt Domsch <Matt_Domsch@dell.com>
* remove check for efi_enabled in exit
* add MODULE_VERSION
*
* 26 Apr 2004 - Matt Domsch <Matt_Domsch@dell.com>
* minor bug fixes
*
* 21 Apr 2004 - Matt Tolentino <matthew.e.tolentino@intel.com)
* converted driver to export variable information via sysfs
* and moved to drivers/firmware directory
* bumped revision number to v0.07 to reflect conversion & move
*
* 10 Dec 2002 - Matt Domsch <Matt_Domsch@dell.com>
* fix locking per Peter Chubb's findings
*
* 25 Mar 2002 - Matt Domsch <Matt_Domsch@dell.com>
* move uuid_unparse() to include/asm-ia64/efi.h:efi_guid_unparse()
*
* 12 Feb 2002 - Matt Domsch <Matt_Domsch@dell.com>
* use list_for_each_safe when deleting vars.
* remove ifdef CONFIG_SMP around include <linux/smp.h>
* v0.04 release to linux-ia64@linuxia64.org
*
* 20 April 2001 - Matt Domsch <Matt_Domsch@dell.com>
* Moved vars from /proc/efi to /proc/efi/vars, and made
* efi.c own the /proc/efi directory.
* v0.03 release to linux-ia64@linuxia64.org
*
* 26 March 2001 - Matt Domsch <Matt_Domsch@dell.com>
* At the request of Stephane, moved ownership of /proc/efi
* to efi.c, and now efivars lives under /proc/efi/vars.
*
* 12 March 2001 - Matt Domsch <Matt_Domsch@dell.com>
* Feedback received from Stephane Eranian incorporated.
* efivar_write() checks copy_from_user() return value.
* efivar_read/write() returns proper errno.
* v0.02 release to linux-ia64@linuxia64.org
*
* 26 February 2001 - Matt Domsch <Matt_Domsch@dell.com>
* v0.01 release to linux-ia64@linuxia64.org
*/
#include <linux/capability.h>
#include <linux/types.h>
#include <linux/errno.h>
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/module.h>
#include <linux/string.h>
#include <linux/smp.h>
#include <linux/efi.h>
#include <linux/sysfs.h>
#include <linux/kobject.h>
#include <linux/device.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
#include <linux/pstore.h>
#include <linux/fs.h>
#include <linux/ramfs.h>
#include <linux/pagemap.h>
#include <asm/uaccess.h>
#define EFIVARS_VERSION "0.08"
#define EFIVARS_DATE "2004-May-17"
MODULE_AUTHOR("Matt Domsch <Matt_Domsch@Dell.com>");
MODULE_DESCRIPTION("sysfs interface to EFI Variables");
MODULE_LICENSE("GPL");
MODULE_VERSION(EFIVARS_VERSION);
#define DUMP_NAME_LEN 52
/*
* Length of a GUID string (strlen("aaaaaaaa-bbbb-cccc-dddd-eeeeeeeeeeee"))
* not including trailing NUL
*/
#define GUID_LEN 36
/*
* The maximum size of VariableName + Data = 1024
* Therefore, it's reasonable to save that much
* space in each part of the structure,
* and we use a page for reading/writing.
*/
struct efi_variable {
efi_char16_t VariableName[1024/sizeof(efi_char16_t)];
efi_guid_t VendorGuid;
unsigned long DataSize;
__u8 Data[1024];
efi_status_t Status;
__u32 Attributes;
} __attribute__((packed));
struct efivar_entry {
struct efivars *efivars;
struct efi_variable var;
struct list_head list;
struct kobject kobj;
};
struct efivar_attribute {
struct attribute attr;
ssize_t (*show) (struct efivar_entry *entry, char *buf);
ssize_t (*store)(struct efivar_entry *entry, const char *buf, size_t count);
};
static struct efivars __efivars;
static struct efivar_operations ops;
#define PSTORE_EFI_ATTRIBUTES \
(EFI_VARIABLE_NON_VOLATILE | \
EFI_VARIABLE_BOOTSERVICE_ACCESS | \
EFI_VARIABLE_RUNTIME_ACCESS)
#define EFIVAR_ATTR(_name, _mode, _show, _store) \
struct efivar_attribute efivar_attr_##_name = { \
.attr = {.name = __stringify(_name), .mode = _mode}, \
.show = _show, \
.store = _store, \
};
#define to_efivar_attr(_attr) container_of(_attr, struct efivar_attribute, attr)
#define to_efivar_entry(obj) container_of(obj, struct efivar_entry, kobj)
/*
* Prototype for sysfs creation function
*/
static int
efivar_create_sysfs_entry(struct efivars *efivars,
unsigned long variable_name_size,
efi_char16_t *variable_name,
efi_guid_t *vendor_guid);
/* Return the number of unicode characters in data */
static unsigned long
utf16_strnlen(efi_char16_t *s, size_t maxlength)
{
unsigned long length = 0;
while (*s++ != 0 && length < maxlength)
length++;
return length;
}
static inline unsigned long
utf16_strlen(efi_char16_t *s)
{
return utf16_strnlen(s, ~0UL);
}
/*
* Return the number of bytes is the length of this string
* Note: this is NOT the same as the number of unicode characters
*/
static inline unsigned long
utf16_strsize(efi_char16_t *data, unsigned long maxlength)
{
return utf16_strnlen(data, maxlength/sizeof(efi_char16_t)) * sizeof(efi_char16_t);
}
static inline int
utf16_strncmp(const efi_char16_t *a, const efi_char16_t *b, size_t len)
{
while (1) {
if (len == 0)
return 0;
if (*a < *b)
return -1;
if (*a > *b)
return 1;
if (*a == 0) /* implies *b == 0 */
return 0;
a++;
b++;
len--;
}
}
static bool
validate_device_path(struct efi_variable *var, int match, u8 *buffer,
unsigned long len)
{
struct efi_generic_dev_path *node;
int offset = 0;
node = (struct efi_generic_dev_path *)buffer;
if (len < sizeof(*node))
return false;
while (offset <= len - sizeof(*node) &&
node->length >= sizeof(*node) &&
node->length <= len - offset) {
offset += node->length;
if ((node->type == EFI_DEV_END_PATH ||
node->type == EFI_DEV_END_PATH2) &&
node->sub_type == EFI_DEV_END_ENTIRE)
return true;
node = (struct efi_generic_dev_path *)(buffer + offset);
}
/*
* If we're here then either node->length pointed past the end
* of the buffer or we reached the end of the buffer without
* finding a device path end node.
*/
return false;
}
static bool
validate_boot_order(struct efi_variable *var, int match, u8 *buffer,
unsigned long len)
{
/* An array of 16-bit integers */
if ((len % 2) != 0)
return false;
return true;
}
static bool
validate_load_option(struct efi_variable *var, int match, u8 *buffer,
unsigned long len)
{
u16 filepathlength;
int i, desclength = 0, namelen;
namelen = utf16_strnlen(var->VariableName, sizeof(var->VariableName));
/* Either "Boot" or "Driver" followed by four digits of hex */
for (i = match; i < match+4; i++) {
if (var->VariableName[i] > 127 ||
hex_to_bin(var->VariableName[i] & 0xff) < 0)
return true;
}
/* Reject it if there's 4 digits of hex and then further content */
if (namelen > match + 4)
return false;
/* A valid entry must be at least 8 bytes */
if (len < 8)
return false;
filepathlength = buffer[4] | buffer[5] << 8;
/*
* There's no stored length for the description, so it has to be
* found by hand
*/
desclength = utf16_strsize((efi_char16_t *)(buffer + 6), len - 6) + 2;
/* Each boot entry must have a descriptor */
if (!desclength)
return false;
/*
* If the sum of the length of the description, the claimed filepath
* length and the original header are greater than the length of the
* variable, it's malformed
*/
if ((desclength + filepathlength + 6) > len)
return false;
/*
* And, finally, check the filepath
*/
return validate_device_path(var, match, buffer + desclength + 6,
filepathlength);
}
static bool
validate_uint16(struct efi_variable *var, int match, u8 *buffer,
unsigned long len)
{
/* A single 16-bit integer */
if (len != 2)
return false;
return true;
}
static bool
validate_ascii_string(struct efi_variable *var, int match, u8 *buffer,
unsigned long len)
{
int i;
for (i = 0; i < len; i++) {
if (buffer[i] > 127)
return false;
if (buffer[i] == 0)
return true;
}
return false;
}
struct variable_validate {
char *name;
bool (*validate)(struct efi_variable *var, int match, u8 *data,
unsigned long len);
};
static const struct variable_validate variable_validate[] = {
{ "BootNext", validate_uint16 },
{ "BootOrder", validate_boot_order },
{ "DriverOrder", validate_boot_order },
{ "Boot*", validate_load_option },
{ "Driver*", validate_load_option },
{ "ConIn", validate_device_path },
{ "ConInDev", validate_device_path },
{ "ConOut", validate_device_path },
{ "ConOutDev", validate_device_path },
{ "ErrOut", validate_device_path },
{ "ErrOutDev", validate_device_path },
{ "Timeout", validate_uint16 },
{ "Lang", validate_ascii_string },
{ "PlatformLang", validate_ascii_string },
{ "", NULL },
};
static bool
validate_var(struct efi_variable *var, u8 *data, unsigned long len)
{
int i;
u16 *unicode_name = var->VariableName;
for (i = 0; variable_validate[i].validate != NULL; i++) {
const char *name = variable_validate[i].name;
int match;
for (match = 0; ; match++) {
char c = name[match];
u16 u = unicode_name[match];
/* All special variables are plain ascii */
if (u > 127)
return true;
/* Wildcard in the matching name means we've matched */
if (c == '*')
return variable_validate[i].validate(var,
match, data, len);
/* Case sensitive match */
if (c != u)
break;
/* Reached the end of the string while matching */
if (!c)
return variable_validate[i].validate(var,
match, data, len);
}
}
return true;
}
static efi_status_t
get_var_data_locked(struct efivars *efivars, struct efi_variable *var)
{
efi_status_t status;
var->DataSize = 1024;
status = efivars->ops->get_variable(var->VariableName,
&var->VendorGuid,
&var->Attributes,
&var->DataSize,
var->Data);
return status;
}
static efi_status_t
get_var_data(struct efivars *efivars, struct efi_variable *var)
{
efi_status_t status;
spin_lock(&efivars->lock);
status = get_var_data_locked(efivars, var);
spin_unlock(&efivars->lock);
if (status != EFI_SUCCESS) {
printk(KERN_WARNING "efivars: get_variable() failed 0x%lx!\n",
status);
}
return status;
}
static ssize_t
efivar_guid_read(struct efivar_entry *entry, char *buf)
{
struct efi_variable *var = &entry->var;
char *str = buf;
if (!entry || !buf)
return 0;
efi_guid_unparse(&var->VendorGuid, str);
str += strlen(str);
str += sprintf(str, "\n");
return str - buf;
}
static ssize_t
efivar_attr_read(struct efivar_entry *entry, char *buf)
{
struct efi_variable *var = &entry->var;
char *str = buf;
efi_status_t status;
if (!entry || !buf)
return -EINVAL;
status = get_var_data(entry->efivars, var);
if (status != EFI_SUCCESS)
return -EIO;
if (var->Attributes & EFI_VARIABLE_NON_VOLATILE)
str += sprintf(str, "EFI_VARIABLE_NON_VOLATILE\n");
if (var->Attributes & EFI_VARIABLE_BOOTSERVICE_ACCESS)
str += sprintf(str, "EFI_VARIABLE_BOOTSERVICE_ACCESS\n");
if (var->Attributes & EFI_VARIABLE_RUNTIME_ACCESS)
str += sprintf(str, "EFI_VARIABLE_RUNTIME_ACCESS\n");
if (var->Attributes & EFI_VARIABLE_HARDWARE_ERROR_RECORD)
str += sprintf(str, "EFI_VARIABLE_HARDWARE_ERROR_RECORD\n");
if (var->Attributes & EFI_VARIABLE_AUTHENTICATED_WRITE_ACCESS)
str += sprintf(str,
"EFI_VARIABLE_AUTHENTICATED_WRITE_ACCESS\n");
if (var->Attributes &
EFI_VARIABLE_TIME_BASED_AUTHENTICATED_WRITE_ACCESS)
str += sprintf(str,
"EFI_VARIABLE_TIME_BASED_AUTHENTICATED_WRITE_ACCESS\n");
if (var->Attributes & EFI_VARIABLE_APPEND_WRITE)
str += sprintf(str, "EFI_VARIABLE_APPEND_WRITE\n");
return str - buf;
}
static ssize_t
efivar_size_read(struct efivar_entry *entry, char *buf)
{
struct efi_variable *var = &entry->var;
char *str = buf;
efi_status_t status;
if (!entry || !buf)
return -EINVAL;
status = get_var_data(entry->efivars, var);
if (status != EFI_SUCCESS)
return -EIO;
str += sprintf(str, "0x%lx\n", var->DataSize);
return str - buf;
}
static ssize_t
efivar_data_read(struct efivar_entry *entry, char *buf)
{
struct efi_variable *var = &entry->var;
efi_status_t status;
if (!entry || !buf)
return -EINVAL;
status = get_var_data(entry->efivars, var);
if (status != EFI_SUCCESS)
return -EIO;
memcpy(buf, var->Data, var->DataSize);
return var->DataSize;
}
/*
* We allow each variable to be edited via rewriting the
* entire efi variable structure.
*/
static ssize_t
efivar_store_raw(struct efivar_entry *entry, const char *buf, size_t count)
{
struct efi_variable *new_var, *var = &entry->var;
struct efivars *efivars = entry->efivars;
efi_status_t status = EFI_NOT_FOUND;
if (count != sizeof(struct efi_variable))
return -EINVAL;
new_var = (struct efi_variable *)buf;
/*
* If only updating the variable data, then the name
* and guid should remain the same
*/
if (memcmp(new_var->VariableName, var->VariableName, sizeof(var->VariableName)) ||
efi_guidcmp(new_var->VendorGuid, var->VendorGuid)) {
printk(KERN_ERR "efivars: Cannot edit the wrong variable!\n");
return -EINVAL;
}
if ((new_var->DataSize <= 0) || (new_var->Attributes == 0)){
printk(KERN_ERR "efivars: DataSize & Attributes must be valid!\n");
return -EINVAL;
}
if ((new_var->Attributes & ~EFI_VARIABLE_MASK) != 0 ||
validate_var(new_var, new_var->Data, new_var->DataSize) == false) {
printk(KERN_ERR "efivars: Malformed variable content\n");
return -EINVAL;
}
spin_lock(&efivars->lock);
status = efivars->ops->set_variable(new_var->VariableName,
&new_var->VendorGuid,
new_var->Attributes,
new_var->DataSize,
new_var->Data);
spin_unlock(&efivars->lock);
if (status != EFI_SUCCESS) {
printk(KERN_WARNING "efivars: set_variable() failed: status=%lx\n",
status);
return -EIO;
}
memcpy(&entry->var, new_var, count);
return count;
}
static ssize_t
efivar_show_raw(struct efivar_entry *entry, char *buf)
{
struct efi_variable *var = &entry->var;
efi_status_t status;
if (!entry || !buf)
return 0;
status = get_var_data(entry->efivars, var);
if (status != EFI_SUCCESS)
return -EIO;
memcpy(buf, var, sizeof(*var));
return sizeof(*var);
}
/*
* Generic read/write functions that call the specific functions of
* the attributes...
*/
static ssize_t efivar_attr_show(struct kobject *kobj, struct attribute *attr,
char *buf)
{
struct efivar_entry *var = to_efivar_entry(kobj);
struct efivar_attribute *efivar_attr = to_efivar_attr(attr);
ssize_t ret = -EIO;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
if (efivar_attr->show) {
ret = efivar_attr->show(var, buf);
}
return ret;
}
static ssize_t efivar_attr_store(struct kobject *kobj, struct attribute *attr,
const char *buf, size_t count)
{
struct efivar_entry *var = to_efivar_entry(kobj);
struct efivar_attribute *efivar_attr = to_efivar_attr(attr);
ssize_t ret = -EIO;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
if (efivar_attr->store)
ret = efivar_attr->store(var, buf, count);
return ret;
}
static const struct sysfs_ops efivar_attr_ops = {
.show = efivar_attr_show,
.store = efivar_attr_store,
};
static void efivar_release(struct kobject *kobj)
{
struct efivar_entry *var = container_of(kobj, struct efivar_entry, kobj);
kfree(var);
}
static EFIVAR_ATTR(guid, 0400, efivar_guid_read, NULL);
static EFIVAR_ATTR(attributes, 0400, efivar_attr_read, NULL);
static EFIVAR_ATTR(size, 0400, efivar_size_read, NULL);
static EFIVAR_ATTR(data, 0400, efivar_data_read, NULL);
static EFIVAR_ATTR(raw_var, 0600, efivar_show_raw, efivar_store_raw);
static struct attribute *def_attrs[] = {
&efivar_attr_guid.attr,
&efivar_attr_size.attr,
&efivar_attr_attributes.attr,
&efivar_attr_data.attr,
&efivar_attr_raw_var.attr,
NULL,
};
static struct kobj_type efivar_ktype = {
.release = efivar_release,
.sysfs_ops = &efivar_attr_ops,
.default_attrs = def_attrs,
};
static inline void
efivar_unregister(struct efivar_entry *var)
{
kobject_put(&var->kobj);
}
static int efivarfs_file_open(struct inode *inode, struct file *file)
{
file->private_data = inode->i_private;
return 0;
}
static int efi_status_to_err(efi_status_t status)
{
int err;
switch (status) {
case EFI_INVALID_PARAMETER:
err = -EINVAL;
break;
case EFI_OUT_OF_RESOURCES:
err = -ENOSPC;
break;
case EFI_DEVICE_ERROR:
err = -EIO;
break;
case EFI_WRITE_PROTECTED:
err = -EROFS;
break;
case EFI_SECURITY_VIOLATION:
err = -EACCES;
break;
case EFI_NOT_FOUND:
err = -EIO;
break;
default:
err = -EINVAL;
}
return err;
}
static ssize_t efivarfs_file_write(struct file *file,
const char __user *userbuf, size_t count, loff_t *ppos)
{
struct efivar_entry *var = file->private_data;
struct efivars *efivars;
efi_status_t status;
void *data;
u32 attributes;
struct inode *inode = file->f_mapping->host;
unsigned long datasize = count - sizeof(attributes);
unsigned long newdatasize;
u64 storage_size, remaining_size, max_size;
ssize_t bytes = 0;
if (count < sizeof(attributes))
return -EINVAL;
if (copy_from_user(&attributes, userbuf, sizeof(attributes)))
return -EFAULT;
if (attributes & ~(EFI_VARIABLE_MASK))
return -EINVAL;
efivars = var->efivars;
/*
* Ensure that the user can't allocate arbitrarily large
* amounts of memory. Pick a default size of 64K if
* QueryVariableInfo() isn't supported by the firmware.
*/
spin_lock(&efivars->lock);
if (!efivars->ops->query_variable_info)
status = EFI_UNSUPPORTED;
else {
const struct efivar_operations *fops = efivars->ops;
status = fops->query_variable_info(attributes, &storage_size,
&remaining_size, &max_size);
}
spin_unlock(&efivars->lock);
if (status != EFI_SUCCESS) {
if (status != EFI_UNSUPPORTED)
return efi_status_to_err(status);
remaining_size = 65536;
}
if (datasize > remaining_size)
return -ENOSPC;
data = kmalloc(datasize, GFP_KERNEL);
if (!data)
return -ENOMEM;
if (copy_from_user(data, userbuf + sizeof(attributes), datasize)) {
bytes = -EFAULT;
goto out;
}
if (validate_var(&var->var, data, datasize) == false) {
bytes = -EINVAL;
goto out;
}
/*
* The lock here protects the get_variable call, the conditional
* set_variable call, and removal of the variable from the efivars
* list (in the case of an authenticated delete).
*/
spin_lock(&efivars->lock);
status = efivars->ops->set_variable(var->var.VariableName,
&var->var.VendorGuid,
attributes, datasize,
data);
if (status != EFI_SUCCESS) {
spin_unlock(&efivars->lock);
kfree(data);
return efi_status_to_err(status);
}
bytes = count;
/*
* Writing to the variable may have caused a change in size (which
* could either be an append or an overwrite), or the variable to be
* deleted. Perform a GetVariable() so we can tell what actually
* happened.
*/
newdatasize = 0;
status = efivars->ops->get_variable(var->var.VariableName,
&var->var.VendorGuid,
NULL, &newdatasize,
NULL);
if (status == EFI_BUFFER_TOO_SMALL) {
spin_unlock(&efivars->lock);
mutex_lock(&inode->i_mutex);
i_size_write(inode, newdatasize + sizeof(attributes));
mutex_unlock(&inode->i_mutex);
} else if (status == EFI_NOT_FOUND) {
list_del(&var->list);
spin_unlock(&efivars->lock);
efivar_unregister(var);
drop_nlink(inode);
efivarfs: Delete dentry from dcache in efivarfs_file_write() Unlike the unlink path that is called from the VFS layer, we need to call d_delete() ourselves when a variable is deleted in efivarfs_file_write(). Failure to do so means we can access a stale struct efivar_entry when reading/writing the file, which can result in the following oops, [ 59.978216] general protection fault: 0000 [#1] SMP [ 60.038660] CPU 9 [ 60.040501] Pid: 1001, comm: cat Not tainted 3.7.0-2.fc19.x86_64 #1 IBM System x3550 M3 -[7944I21]-/69Y4438 [ 60.050840] RIP: 0010:[<ffffffff810d5d1e>] [<ffffffff810d5d1e>] __lock_acquire+0x5e/0x1bb0 [ 60.059198] RSP: 0018:ffff880270595ce8 EFLAGS: 00010046 [ 60.064500] RAX: 0000000000000046 RBX: 0000000000000002 RCX: 0000000000000000 [ 60.071617] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 6b6b6b6b6b6b6b83 [ 60.078735] RBP: ffff880270595dd8 R08: 0000000000000002 R09: 0000000000000000 [ 60.085852] R10: 6b6b6b6b6b6b6b83 R11: 0000000000000000 R12: 0000000000000000 [ 60.092971] R13: ffff88027170cd20 R14: 0000000000000000 R15: 0000000000000000 [ 60.100091] FS: 00007fc0c8ff3740(0000) GS:ffff880277000000(0000) knlGS:0000000000000000 [ 60.108164] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 60.113899] CR2: 0000000001520000 CR3: 000000026d594000 CR4: 00000000000007e0 [ 60.121016] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 60.128135] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [ 60.135254] Process cat (pid: 1001, threadinfo ffff880270594000, task ffff88027170cd20) [ 60.143239] Stack: [ 60.145251] ffff880270595cf8 ffffffff81021da3 ffff880270595d08 ffffffff81021e19 [ 60.152714] ffff880270595d38 ffffffff810acdb5 ffff880200000168 0000000000000086 [ 60.160175] ffff88027170d5e8 ffffffff810d25ed ffff880270595d58 ffffffff810ace7f [ 60.167638] Call Trace: [ 60.170088] [<ffffffff81021da3>] ? native_sched_clock+0x13/0x80 [ 60.176085] [<ffffffff81021e19>] ? sched_clock+0x9/0x10 [ 60.181389] [<ffffffff810acdb5>] ? sched_clock_cpu+0xc5/0x120 [ 60.187211] [<ffffffff810d25ed>] ? trace_hardirqs_off+0xd/0x10 [ 60.193121] [<ffffffff810ace7f>] ? local_clock+0x6f/0x80 [ 60.198513] [<ffffffff810d2f6f>] ? lock_release_holdtime.part.26+0xf/0x180 [ 60.205465] [<ffffffff810d7b57>] ? lock_release_non_nested+0x2e7/0x320 [ 60.212073] [<ffffffff815638bb>] ? efivarfs_file_write+0x5b/0x280 [ 60.218242] [<ffffffff810d7f41>] lock_acquire+0xa1/0x1f0 [ 60.223633] [<ffffffff81563971>] ? efivarfs_file_write+0x111/0x280 [ 60.229892] [<ffffffff8118b47c>] ? might_fault+0x5c/0xb0 [ 60.235287] [<ffffffff816f1bf6>] _raw_spin_lock+0x46/0x80 [ 60.240762] [<ffffffff81563971>] ? efivarfs_file_write+0x111/0x280 [ 60.247018] [<ffffffff81563971>] efivarfs_file_write+0x111/0x280 [ 60.253103] [<ffffffff811d307f>] vfs_write+0xaf/0x190 [ 60.258233] [<ffffffff811d33d5>] sys_write+0x55/0xa0 [ 60.263278] [<ffffffff816fbd19>] system_call_fastpath+0x16/0x1b [ 60.269271] Code: 41 0f 45 d8 4c 89 75 f0 4c 89 7d f8 85 c0 0f 84 09 01 00 00 8b 05 a3 f9 ff 00 49 89 fa 41 89 f6 41 89 d3 85 c0 0f 84 12 01 00 00 <49> 8b 02 ba 01 00 00 00 48 3d a0 07 14 82 0f 44 da 41 83 fe 01 [ 60.289431] RIP [<ffffffff810d5d1e>] __lock_acquire+0x5e/0x1bb0 [ 60.295444] RSP <ffff880270595ce8> [ 60.298928] ---[ end trace 1bbfd41a2cf6a0d8 ]--- Cc: Josh Boyer <jwboyer@redhat.com> Acked-by: Jeremy Kerr <jeremy.kerr@canonical.com> Cc: Lee, Chun-Yi <jlee@suse.com> Cc: Andy Whitcroft <apw@canonical.com> Reported-by: Lingzhu Xiang <lxiang@redhat.com> Signed-off-by: Matt Fleming <matt.fleming@intel.com>
2013-01-16 21:55:36 +00:00
d_delete(file->f_dentry);
dput(file->f_dentry);
} else {
spin_unlock(&efivars->lock);
pr_warn("efivarfs: inconsistent EFI variable implementation? "
"status = %lx\n", status);
}
out:
kfree(data);
return bytes;
}
static ssize_t efivarfs_file_read(struct file *file, char __user *userbuf,
size_t count, loff_t *ppos)
{
struct efivar_entry *var = file->private_data;
struct efivars *efivars = var->efivars;
efi_status_t status;
unsigned long datasize = 0;
u32 attributes;
void *data;
ssize_t size = 0;
spin_lock(&efivars->lock);
status = efivars->ops->get_variable(var->var.VariableName,
&var->var.VendorGuid,
&attributes, &datasize, NULL);
spin_unlock(&efivars->lock);
if (status != EFI_BUFFER_TOO_SMALL)
return efi_status_to_err(status);
data = kmalloc(datasize + sizeof(attributes), GFP_KERNEL);
if (!data)
return -ENOMEM;
spin_lock(&efivars->lock);
status = efivars->ops->get_variable(var->var.VariableName,
&var->var.VendorGuid,
&attributes, &datasize,
(data + sizeof(attributes)));
spin_unlock(&efivars->lock);
if (status != EFI_SUCCESS) {
size = efi_status_to_err(status);
goto out_free;
}
memcpy(data, &attributes, sizeof(attributes));
size = simple_read_from_buffer(userbuf, count, ppos,
data, datasize + sizeof(attributes));
out_free:
kfree(data);
return size;
}
static void efivarfs_evict_inode(struct inode *inode)
{
clear_inode(inode);
}
static const struct super_operations efivarfs_ops = {
.statfs = simple_statfs,
.drop_inode = generic_delete_inode,
.evict_inode = efivarfs_evict_inode,
.show_options = generic_show_options,
};
static struct super_block *efivarfs_sb;
static const struct inode_operations efivarfs_dir_inode_operations;
static const struct file_operations efivarfs_file_operations = {
.open = efivarfs_file_open,
.read = efivarfs_file_read,
.write = efivarfs_file_write,
.llseek = no_llseek,
};
static struct inode *efivarfs_get_inode(struct super_block *sb,
const struct inode *dir, int mode, dev_t dev)
{
struct inode *inode = new_inode(sb);
if (inode) {
inode->i_ino = get_next_ino();
inode->i_mode = mode;
inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
switch (mode & S_IFMT) {
case S_IFREG:
inode->i_fop = &efivarfs_file_operations;
break;
case S_IFDIR:
inode->i_op = &efivarfs_dir_inode_operations;
inode->i_fop = &simple_dir_operations;
inc_nlink(inode);
break;
}
}
return inode;
}
static void efivarfs_hex_to_guid(const char *str, efi_guid_t *guid)
{
guid->b[0] = hex_to_bin(str[6]) << 4 | hex_to_bin(str[7]);
guid->b[1] = hex_to_bin(str[4]) << 4 | hex_to_bin(str[5]);
guid->b[2] = hex_to_bin(str[2]) << 4 | hex_to_bin(str[3]);
guid->b[3] = hex_to_bin(str[0]) << 4 | hex_to_bin(str[1]);
guid->b[4] = hex_to_bin(str[11]) << 4 | hex_to_bin(str[12]);
guid->b[5] = hex_to_bin(str[9]) << 4 | hex_to_bin(str[10]);
guid->b[6] = hex_to_bin(str[16]) << 4 | hex_to_bin(str[17]);
guid->b[7] = hex_to_bin(str[14]) << 4 | hex_to_bin(str[15]);
guid->b[8] = hex_to_bin(str[19]) << 4 | hex_to_bin(str[20]);
guid->b[9] = hex_to_bin(str[21]) << 4 | hex_to_bin(str[22]);
guid->b[10] = hex_to_bin(str[24]) << 4 | hex_to_bin(str[25]);
guid->b[11] = hex_to_bin(str[26]) << 4 | hex_to_bin(str[27]);
guid->b[12] = hex_to_bin(str[28]) << 4 | hex_to_bin(str[29]);
guid->b[13] = hex_to_bin(str[30]) << 4 | hex_to_bin(str[31]);
guid->b[14] = hex_to_bin(str[32]) << 4 | hex_to_bin(str[33]);
guid->b[15] = hex_to_bin(str[34]) << 4 | hex_to_bin(str[35]);
}
static int efivarfs_create(struct inode *dir, struct dentry *dentry,
umode_t mode, bool excl)
{
struct inode *inode;
struct efivars *efivars = &__efivars;
struct efivar_entry *var;
int namelen, i = 0, err = 0;
/*
* We need a GUID, plus at least one letter for the variable name,
* plus the '-' separator
*/
if (dentry->d_name.len < GUID_LEN + 2)
return -EINVAL;
inode = efivarfs_get_inode(dir->i_sb, dir, mode, 0);
if (!inode)
return -ENOMEM;
var = kzalloc(sizeof(struct efivar_entry), GFP_KERNEL);
if (!var) {
err = -ENOMEM;
goto out;
}
/* length of the variable name itself: remove GUID and separator */
namelen = dentry->d_name.len - GUID_LEN - 1;
efivarfs_hex_to_guid(dentry->d_name.name + namelen + 1,
&var->var.VendorGuid);
for (i = 0; i < namelen; i++)
var->var.VariableName[i] = dentry->d_name.name[i];
var->var.VariableName[i] = '\0';
inode->i_private = var;
var->efivars = efivars;
var->kobj.kset = efivars->kset;
err = kobject_init_and_add(&var->kobj, &efivar_ktype, NULL, "%s",
dentry->d_name.name);
if (err)
goto out;
kobject_uevent(&var->kobj, KOBJ_ADD);
spin_lock(&efivars->lock);
list_add(&var->list, &efivars->list);
spin_unlock(&efivars->lock);
d_instantiate(dentry, inode);
dget(dentry);
out:
if (err) {
kfree(var);
iput(inode);
}
return err;
}
static int efivarfs_unlink(struct inode *dir, struct dentry *dentry)
{
struct efivar_entry *var = dentry->d_inode->i_private;
struct efivars *efivars = var->efivars;
efi_status_t status;
spin_lock(&efivars->lock);
status = efivars->ops->set_variable(var->var.VariableName,
&var->var.VendorGuid,
0, 0, NULL);
if (status == EFI_SUCCESS || status == EFI_NOT_FOUND) {
list_del(&var->list);
spin_unlock(&efivars->lock);
efivar_unregister(var);
drop_nlink(dentry->d_inode);
dput(dentry);
return 0;
}
spin_unlock(&efivars->lock);
return -EINVAL;
};
static int efivarfs_fill_super(struct super_block *sb, void *data, int silent)
{
struct inode *inode = NULL;
struct dentry *root;
struct efivar_entry *entry, *n;
struct efivars *efivars = &__efivars;
char *name;
efivarfs_sb = sb;
sb->s_maxbytes = MAX_LFS_FILESIZE;
sb->s_blocksize = PAGE_CACHE_SIZE;
sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
sb->s_magic = EFIVARFS_MAGIC;
sb->s_op = &efivarfs_ops;
sb->s_time_gran = 1;
inode = efivarfs_get_inode(sb, NULL, S_IFDIR | 0755, 0);
if (!inode)
return -ENOMEM;
inode->i_op = &efivarfs_dir_inode_operations;
root = d_make_root(inode);
sb->s_root = root;
if (!root)
return -ENOMEM;
list_for_each_entry_safe(entry, n, &efivars->list, list) {
struct dentry *dentry, *root = efivarfs_sb->s_root;
unsigned long size = 0;
int len, i;
inode = NULL;
len = utf16_strlen(entry->var.VariableName);
/* name, plus '-', plus GUID, plus NUL*/
name = kmalloc(len + 1 + GUID_LEN + 1, GFP_ATOMIC);
if (!name)
goto fail;
for (i = 0; i < len; i++)
name[i] = entry->var.VariableName[i] & 0xFF;
name[len] = '-';
efi_guid_unparse(&entry->var.VendorGuid, name + len + 1);
name[len+GUID_LEN+1] = '\0';
inode = efivarfs_get_inode(efivarfs_sb, root->d_inode,
S_IFREG | 0644, 0);
if (!inode)
goto fail_name;
dentry = d_alloc_name(root, name);
if (!dentry)
goto fail_inode;
/* copied by the above to local storage in the dentry. */
kfree(name);
spin_lock(&efivars->lock);
efivars->ops->get_variable(entry->var.VariableName,
&entry->var.VendorGuid,
&entry->var.Attributes,
&size,
NULL);
spin_unlock(&efivars->lock);
mutex_lock(&inode->i_mutex);
inode->i_private = entry;
i_size_write(inode, size+4);
mutex_unlock(&inode->i_mutex);
d_add(dentry, inode);
}
return 0;
fail_inode:
iput(inode);
fail_name:
kfree(name);
fail:
return -ENOMEM;
}
static struct dentry *efivarfs_mount(struct file_system_type *fs_type,
int flags, const char *dev_name, void *data)
{
return mount_single(fs_type, flags, data, efivarfs_fill_super);
}
static void efivarfs_kill_sb(struct super_block *sb)
{
kill_litter_super(sb);
efivarfs_sb = NULL;
}
static struct file_system_type efivarfs_type = {
.name = "efivarfs",
.mount = efivarfs_mount,
.kill_sb = efivarfs_kill_sb,
};
static const struct inode_operations efivarfs_dir_inode_operations = {
.lookup = simple_lookup,
.unlink = efivarfs_unlink,
.create = efivarfs_create,
};
static struct pstore_info efi_pstore_info;
#ifdef CONFIG_PSTORE
static int efi_pstore_open(struct pstore_info *psi)
{
struct efivars *efivars = psi->data;
spin_lock(&efivars->lock);
efivars->walk_entry = list_first_entry(&efivars->list,
struct efivar_entry, list);
return 0;
}
static int efi_pstore_close(struct pstore_info *psi)
{
struct efivars *efivars = psi->data;
spin_unlock(&efivars->lock);
return 0;
}
static ssize_t efi_pstore_read(u64 *id, enum pstore_type_id *type,
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
int *count, struct timespec *timespec,
pstore: pass allocated memory region back to caller The buf_lock cannot be held while populating the inodes, so make the backend pass forward an allocated and filled buffer instead. This solves the following backtrace. The effect is that "buf" is only ever used to notify the backends that something was written to it, and shouldn't be used in the read path. To replace the buf_lock during the read path, isolate the open/read/close loop with a separate mutex to maintain serialized access to the backend. Note that is is up to the pstore backend to cope if the (*write)() path is called in the middle of the read path. [ 59.691019] BUG: sleeping function called from invalid context at .../mm/slub.c:847 [ 59.691019] in_atomic(): 0, irqs_disabled(): 1, pid: 1819, name: mount [ 59.691019] Pid: 1819, comm: mount Not tainted 3.0.8 #1 [ 59.691019] Call Trace: [ 59.691019] [<810252d5>] __might_sleep+0xc3/0xca [ 59.691019] [<810a26e6>] kmem_cache_alloc+0x32/0xf3 [ 59.691019] [<810b53ac>] ? __d_lookup_rcu+0x6f/0xf4 [ 59.691019] [<810b68b1>] alloc_inode+0x2a/0x64 [ 59.691019] [<810b6903>] new_inode+0x18/0x43 [ 59.691019] [<81142447>] pstore_get_inode.isra.1+0x11/0x98 [ 59.691019] [<81142623>] pstore_mkfile+0xae/0x26f [ 59.691019] [<810a2a66>] ? kmem_cache_free+0x19/0xb1 [ 59.691019] [<8116c821>] ? ida_get_new_above+0x140/0x158 [ 59.691019] [<811708ea>] ? __init_rwsem+0x1e/0x2c [ 59.691019] [<810b67e8>] ? inode_init_always+0x111/0x1b0 [ 59.691019] [<8102127e>] ? should_resched+0xd/0x27 [ 59.691019] [<8137977f>] ? _cond_resched+0xd/0x21 [ 59.691019] [<81142abf>] pstore_get_records+0x52/0xa7 [ 59.691019] [<8114254b>] pstore_fill_super+0x7d/0x91 [ 59.691019] [<810a7ff5>] mount_single+0x46/0x82 [ 59.691019] [<8114231a>] pstore_mount+0x15/0x17 [ 59.691019] [<811424ce>] ? pstore_get_inode.isra.1+0x98/0x98 [ 59.691019] [<810a8199>] mount_fs+0x5a/0x12d [ 59.691019] [<810b9174>] ? alloc_vfsmnt+0xa4/0x14a [ 59.691019] [<810b9474>] vfs_kern_mount+0x4f/0x7d [ 59.691019] [<810b9d7e>] do_kern_mount+0x34/0xb2 [ 59.691019] [<810bb15f>] do_mount+0x5fc/0x64a [ 59.691019] [<810912fb>] ? strndup_user+0x2e/0x3f [ 59.691019] [<810bb3cb>] sys_mount+0x66/0x99 [ 59.691019] [<8137b537>] sysenter_do_call+0x12/0x26 Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Tony Luck <tony.luck@intel.com>
2011-11-17 20:58:07 +00:00
char **buf, struct pstore_info *psi)
{
efi_guid_t vendor = LINUX_EFI_CRASH_GUID;
struct efivars *efivars = psi->data;
char name[DUMP_NAME_LEN];
int i;
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
int cnt;
unsigned int part, size;
unsigned long time;
while (&efivars->walk_entry->list != &efivars->list) {
if (!efi_guidcmp(efivars->walk_entry->var.VendorGuid,
vendor)) {
for (i = 0; i < DUMP_NAME_LEN; i++) {
name[i] = efivars->walk_entry->var.VariableName[i];
}
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
if (sscanf(name, "dump-type%u-%u-%d-%lu",
type, &part, &cnt, &time) == 4) {
*id = part;
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
*count = cnt;
timespec->tv_sec = time;
timespec->tv_nsec = 0;
} else if (sscanf(name, "dump-type%u-%u-%lu",
type, &part, &time) == 3) {
/*
* Check if an old format,
* which doesn't support holding
* multiple logs, remains.
*/
*id = part;
*count = 0;
timespec->tv_sec = time;
timespec->tv_nsec = 0;
} else {
efivars->walk_entry = list_entry(
efivars->walk_entry->list.next,
struct efivar_entry, list);
continue;
}
get_var_data_locked(efivars, &efivars->walk_entry->var);
size = efivars->walk_entry->var.DataSize;
*buf = kmalloc(size, GFP_KERNEL);
if (*buf == NULL)
return -ENOMEM;
memcpy(*buf, efivars->walk_entry->var.Data,
size);
efivars->walk_entry = list_entry(
efivars->walk_entry->list.next,
struct efivar_entry, list);
return size;
}
efivars->walk_entry = list_entry(efivars->walk_entry->list.next,
struct efivar_entry, list);
}
return 0;
}
static int efi_pstore_write(enum pstore_type_id type,
enum kmsg_dump_reason reason, u64 *id,
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
unsigned int part, int count, size_t size,
struct pstore_info *psi)
{
char name[DUMP_NAME_LEN];
efi_char16_t efi_name[DUMP_NAME_LEN];
efi_guid_t vendor = LINUX_EFI_CRASH_GUID;
struct efivars *efivars = psi->data;
int i, ret = 0;
u64 storage_space, remaining_space, max_variable_size;
efi_status_t status = EFI_NOT_FOUND;
spin_lock(&efivars->lock);
/*
* Check if there is a space enough to log.
* size: a size of logging data
* DUMP_NAME_LEN * 2: a maximum size of variable name
*/
status = efivars->ops->query_variable_info(PSTORE_EFI_ATTRIBUTES,
&storage_space,
&remaining_space,
&max_variable_size);
if (status || remaining_space < size + DUMP_NAME_LEN * 2) {
spin_unlock(&efivars->lock);
*id = part;
return -ENOSPC;
}
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
sprintf(name, "dump-type%u-%u-%d-%lu", type, part, count,
get_seconds());
for (i = 0; i < DUMP_NAME_LEN; i++)
efi_name[i] = name[i];
efivars->ops->set_variable(efi_name, &vendor, PSTORE_EFI_ATTRIBUTES,
size, psi->buf);
spin_unlock(&efivars->lock);
if (size)
ret = efivar_create_sysfs_entry(efivars,
utf16_strsize(efi_name,
DUMP_NAME_LEN * 2),
efi_name, &vendor);
*id = part;
return ret;
};
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
static int efi_pstore_erase(enum pstore_type_id type, u64 id, int count,
struct timespec time, struct pstore_info *psi)
{
char name[DUMP_NAME_LEN];
efi_char16_t efi_name[DUMP_NAME_LEN];
char name_old[DUMP_NAME_LEN];
efi_char16_t efi_name_old[DUMP_NAME_LEN];
efi_guid_t vendor = LINUX_EFI_CRASH_GUID;
struct efivars *efivars = psi->data;
struct efivar_entry *entry, *found = NULL;
int i;
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
sprintf(name, "dump-type%u-%u-%d-%lu", type, (unsigned int)id, count,
time.tv_sec);
spin_lock(&efivars->lock);
for (i = 0; i < DUMP_NAME_LEN; i++)
efi_name[i] = name[i];
/*
* Clean up an entry with the same name
*/
list_for_each_entry(entry, &efivars->list, list) {
get_var_data_locked(efivars, &entry->var);
if (efi_guidcmp(entry->var.VendorGuid, vendor))
continue;
if (utf16_strncmp(entry->var.VariableName, efi_name,
utf16_strlen(efi_name))) {
/*
* Check if an old format,
* which doesn't support holding
* multiple logs, remains.
*/
sprintf(name_old, "dump-type%u-%u-%lu", type,
(unsigned int)id, time.tv_sec);
for (i = 0; i < DUMP_NAME_LEN; i++)
efi_name_old[i] = name_old[i];
if (utf16_strncmp(entry->var.VariableName, efi_name_old,
utf16_strlen(efi_name_old)))
continue;
}
/* found */
found = entry;
efivars->ops->set_variable(entry->var.VariableName,
&entry->var.VendorGuid,
PSTORE_EFI_ATTRIBUTES,
0, NULL);
break;
}
if (found)
list_del(&found->list);
spin_unlock(&efivars->lock);
if (found)
efivar_unregister(found);
return 0;
}
#else
static int efi_pstore_open(struct pstore_info *psi)
{
return 0;
}
static int efi_pstore_close(struct pstore_info *psi)
{
return 0;
}
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
static ssize_t efi_pstore_read(u64 *id, enum pstore_type_id *type, int *count,
struct timespec *timespec,
char **buf, struct pstore_info *psi)
{
return -1;
}
static int efi_pstore_write(enum pstore_type_id type,
enum kmsg_dump_reason reason, u64 *id,
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
unsigned int part, int count, size_t size,
struct pstore_info *psi)
{
return 0;
}
efi_pstore: Add a sequence counter to a variable name [Issue] Currently, a variable name, which identifies each entry, consists of type, id and ctime. But if multiple events happens in a short time, a second/third event may fail to log because efi_pstore can't distinguish each event with current variable name. [Solution] A reasonable way to identify all events precisely is introducing a sequence counter to the variable name. The sequence counter has already supported in a pstore layer with "oopscount". So, this patch adds it to a variable name. Also, it is passed to read/erase callbacks of platform drivers in accordance with the modification of the variable name. <before applying this patch> a variable name of first event: dump-type0-1-12345678 a variable name of second event: dump-type0-1-12345678 type:0 id:1 ctime:12345678 If multiple events happen in a short time, efi_pstore can't distinguish them because variable names are same among them. <after applying this patch> it can be distinguishable by adding a sequence counter as follows. a variable name of first event: dump-type0-1-1-12345678 a variable name of Second event: dump-type0-1-2-12345678 type:0 id:1 sequence counter: 1(first event), 2(second event) ctime:12345678 In case of a write callback executed in pstore_console_write(), "0" is added to an argument of the write callback because it just logs all kernel messages and doesn't need to care about multiple events. Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Mike Waychison <mikew@google.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2012-11-27 00:07:44 +00:00
static int efi_pstore_erase(enum pstore_type_id type, u64 id, int count,
struct timespec time, struct pstore_info *psi)
{
return 0;
}
#endif
static struct pstore_info efi_pstore_info = {
.owner = THIS_MODULE,
.name = "efi",
.open = efi_pstore_open,
.close = efi_pstore_close,
.read = efi_pstore_read,
.write = efi_pstore_write,
.erase = efi_pstore_erase,
};
static ssize_t efivar_create(struct file *filp, struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t pos, size_t count)
{
struct efi_variable *new_var = (struct efi_variable *)buf;
struct efivars *efivars = bin_attr->private;
struct efivar_entry *search_efivar, *n;
unsigned long strsize1, strsize2;
efi_status_t status = EFI_NOT_FOUND;
int found = 0;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
if ((new_var->Attributes & ~EFI_VARIABLE_MASK) != 0 ||
validate_var(new_var, new_var->Data, new_var->DataSize) == false) {
printk(KERN_ERR "efivars: Malformed variable content\n");
return -EINVAL;
}
spin_lock(&efivars->lock);
/*
* Does this variable already exist?
*/
list_for_each_entry_safe(search_efivar, n, &efivars->list, list) {
strsize1 = utf16_strsize(search_efivar->var.VariableName, 1024);
strsize2 = utf16_strsize(new_var->VariableName, 1024);
if (strsize1 == strsize2 &&
!memcmp(&(search_efivar->var.VariableName),
new_var->VariableName, strsize1) &&
!efi_guidcmp(search_efivar->var.VendorGuid,
new_var->VendorGuid)) {
found = 1;
break;
}
}
if (found) {
spin_unlock(&efivars->lock);
return -EINVAL;
}
/* now *really* create the variable via EFI */
status = efivars->ops->set_variable(new_var->VariableName,
&new_var->VendorGuid,
new_var->Attributes,
new_var->DataSize,
new_var->Data);
if (status != EFI_SUCCESS) {
printk(KERN_WARNING "efivars: set_variable() failed: status=%lx\n",
status);
spin_unlock(&efivars->lock);
return -EIO;
}
spin_unlock(&efivars->lock);
/* Create the entry in sysfs. Locking is not required here */
status = efivar_create_sysfs_entry(efivars,
utf16_strsize(new_var->VariableName,
1024),
new_var->VariableName,
&new_var->VendorGuid);
if (status) {
printk(KERN_WARNING "efivars: variable created, but sysfs entry wasn't.\n");
}
return count;
}
static ssize_t efivar_delete(struct file *filp, struct kobject *kobj,
struct bin_attribute *bin_attr,
char *buf, loff_t pos, size_t count)
{
struct efi_variable *del_var = (struct efi_variable *)buf;
struct efivars *efivars = bin_attr->private;
struct efivar_entry *search_efivar, *n;
unsigned long strsize1, strsize2;
efi_status_t status = EFI_NOT_FOUND;
int found = 0;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
spin_lock(&efivars->lock);
/*
* Does this variable already exist?
*/
list_for_each_entry_safe(search_efivar, n, &efivars->list, list) {
strsize1 = utf16_strsize(search_efivar->var.VariableName, 1024);
strsize2 = utf16_strsize(del_var->VariableName, 1024);
if (strsize1 == strsize2 &&
!memcmp(&(search_efivar->var.VariableName),
del_var->VariableName, strsize1) &&
!efi_guidcmp(search_efivar->var.VendorGuid,
del_var->VendorGuid)) {
found = 1;
break;
}
}
if (!found) {
spin_unlock(&efivars->lock);
return -EINVAL;
}
/* force the Attributes/DataSize to 0 to ensure deletion */
del_var->Attributes = 0;
del_var->DataSize = 0;
status = efivars->ops->set_variable(del_var->VariableName,
&del_var->VendorGuid,
del_var->Attributes,
del_var->DataSize,
del_var->Data);
if (status != EFI_SUCCESS) {
printk(KERN_WARNING "efivars: set_variable() failed: status=%lx\n",
status);
spin_unlock(&efivars->lock);
return -EIO;
}
list_del(&search_efivar->list);
/* We need to release this lock before unregistering. */
spin_unlock(&efivars->lock);
efivar_unregister(search_efivar);
/* It's dead Jim.... */
return count;
}
/*
* Let's not leave out systab information that snuck into
* the efivars driver
*/
static ssize_t systab_show(struct kobject *kobj,
struct kobj_attribute *attr, char *buf)
{
char *str = buf;
if (!kobj || !buf)
return -EINVAL;
if (efi.mps != EFI_INVALID_TABLE_ADDR)
str += sprintf(str, "MPS=0x%lx\n", efi.mps);
if (efi.acpi20 != EFI_INVALID_TABLE_ADDR)
str += sprintf(str, "ACPI20=0x%lx\n", efi.acpi20);
if (efi.acpi != EFI_INVALID_TABLE_ADDR)
str += sprintf(str, "ACPI=0x%lx\n", efi.acpi);
if (efi.smbios != EFI_INVALID_TABLE_ADDR)
str += sprintf(str, "SMBIOS=0x%lx\n", efi.smbios);
if (efi.hcdp != EFI_INVALID_TABLE_ADDR)
str += sprintf(str, "HCDP=0x%lx\n", efi.hcdp);
if (efi.boot_info != EFI_INVALID_TABLE_ADDR)
str += sprintf(str, "BOOTINFO=0x%lx\n", efi.boot_info);
if (efi.uga != EFI_INVALID_TABLE_ADDR)
str += sprintf(str, "UGA=0x%lx\n", efi.uga);
return str - buf;
}
static struct kobj_attribute efi_attr_systab =
__ATTR(systab, 0400, systab_show, NULL);
static struct attribute *efi_subsys_attrs[] = {
&efi_attr_systab.attr,
NULL, /* maybe more in the future? */
};
static struct attribute_group efi_subsys_attr_group = {
.attrs = efi_subsys_attrs,
};
static struct kobject *efi_kobj;
/*
* efivar_create_sysfs_entry()
* Requires:
* variable_name_size = number of bytes required to hold
* variable_name (not counting the NULL
* character at the end.
* efivars->lock is not held on entry or exit.
* Returns 1 on failure, 0 on success
*/
static int
efivar_create_sysfs_entry(struct efivars *efivars,
unsigned long variable_name_size,
efi_char16_t *variable_name,
efi_guid_t *vendor_guid)
{
int i, short_name_size;
char *short_name;
struct efivar_entry *new_efivar;
/*
* Length of the variable bytes in ASCII, plus the '-' separator,
* plus the GUID, plus trailing NUL
*/
short_name_size = variable_name_size / sizeof(efi_char16_t)
+ 1 + GUID_LEN + 1;
short_name = kzalloc(short_name_size, GFP_KERNEL);
new_efivar = kzalloc(sizeof(struct efivar_entry), GFP_KERNEL);
if (!short_name || !new_efivar) {
kfree(short_name);
kfree(new_efivar);
return 1;
}
new_efivar->efivars = efivars;
memcpy(new_efivar->var.VariableName, variable_name,
variable_name_size);
memcpy(&(new_efivar->var.VendorGuid), vendor_guid, sizeof(efi_guid_t));
/* Convert Unicode to normal chars (assume top bits are 0),
ala UTF-8 */
for (i=0; i < (int)(variable_name_size / sizeof(efi_char16_t)); i++) {
short_name[i] = variable_name[i] & 0xFF;
}
/* This is ugly, but necessary to separate one vendor's
private variables from another's. */
*(short_name + strlen(short_name)) = '-';
efi_guid_unparse(vendor_guid, short_name + strlen(short_name));
new_efivar->kobj.kset = efivars->kset;
i = kobject_init_and_add(&new_efivar->kobj, &efivar_ktype, NULL,
"%s", short_name);
if (i) {
kfree(short_name);
kfree(new_efivar);
return 1;
}
kobject_uevent(&new_efivar->kobj, KOBJ_ADD);
kfree(short_name);
short_name = NULL;
spin_lock(&efivars->lock);
list_add(&new_efivar->list, &efivars->list);
spin_unlock(&efivars->lock);
return 0;
}
static int
create_efivars_bin_attributes(struct efivars *efivars)
{
struct bin_attribute *attr;
int error;
/* new_var */
attr = kzalloc(sizeof(*attr), GFP_KERNEL);
if (!attr)
return -ENOMEM;
attr->attr.name = "new_var";
attr->attr.mode = 0200;
attr->write = efivar_create;
attr->private = efivars;
efivars->new_var = attr;
/* del_var */
attr = kzalloc(sizeof(*attr), GFP_KERNEL);
if (!attr) {
error = -ENOMEM;
goto out_free;
}
attr->attr.name = "del_var";
attr->attr.mode = 0200;
attr->write = efivar_delete;
attr->private = efivars;
efivars->del_var = attr;
sysfs_bin_attr_init(efivars->new_var);
sysfs_bin_attr_init(efivars->del_var);
/* Register */
error = sysfs_create_bin_file(&efivars->kset->kobj,
efivars->new_var);
if (error) {
printk(KERN_ERR "efivars: unable to create new_var sysfs file"
" due to error %d\n", error);
goto out_free;
}
error = sysfs_create_bin_file(&efivars->kset->kobj,
efivars->del_var);
if (error) {
printk(KERN_ERR "efivars: unable to create del_var sysfs file"
" due to error %d\n", error);
sysfs_remove_bin_file(&efivars->kset->kobj,
efivars->new_var);
goto out_free;
}
return 0;
out_free:
kfree(efivars->del_var);
efivars->del_var = NULL;
kfree(efivars->new_var);
efivars->new_var = NULL;
return error;
}
void unregister_efivars(struct efivars *efivars)
{
struct efivar_entry *entry, *n;
list_for_each_entry_safe(entry, n, &efivars->list, list) {
spin_lock(&efivars->lock);
list_del(&entry->list);
spin_unlock(&efivars->lock);
efivar_unregister(entry);
}
if (efivars->new_var)
sysfs_remove_bin_file(&efivars->kset->kobj, efivars->new_var);
if (efivars->del_var)
sysfs_remove_bin_file(&efivars->kset->kobj, efivars->del_var);
kfree(efivars->new_var);
kfree(efivars->del_var);
kobject_put(efivars->kobject);
kset_unregister(efivars->kset);
}
EXPORT_SYMBOL_GPL(unregister_efivars);
int register_efivars(struct efivars *efivars,
const struct efivar_operations *ops,
struct kobject *parent_kobj)
{
efi_status_t status = EFI_NOT_FOUND;
efi_guid_t vendor_guid;
efi_char16_t *variable_name;
unsigned long variable_name_size = 1024;
int error = 0;
variable_name = kzalloc(variable_name_size, GFP_KERNEL);
if (!variable_name) {
printk(KERN_ERR "efivars: Memory allocation failed.\n");
return -ENOMEM;
}
spin_lock_init(&efivars->lock);
INIT_LIST_HEAD(&efivars->list);
efivars->ops = ops;
efivars->kset = kset_create_and_add("vars", NULL, parent_kobj);
if (!efivars->kset) {
printk(KERN_ERR "efivars: Subsystem registration failed.\n");
error = -ENOMEM;
goto out;
}
efivars->kobject = kobject_create_and_add("efivars", parent_kobj);
if (!efivars->kobject) {
pr_err("efivars: Subsystem registration failed.\n");
error = -ENOMEM;
kset_unregister(efivars->kset);
goto out;
}
/*
* Per EFI spec, the maximum storage allocated for both
* the variable name and variable data is 1024 bytes.
*/
do {
variable_name_size = 1024;
status = ops->get_next_variable(&variable_name_size,
variable_name,
&vendor_guid);
switch (status) {
case EFI_SUCCESS:
efivar_create_sysfs_entry(efivars,
variable_name_size,
variable_name,
&vendor_guid);
break;
case EFI_NOT_FOUND:
break;
default:
printk(KERN_WARNING "efivars: get_next_variable: status=%lx\n",
status);
status = EFI_NOT_FOUND;
break;
}
} while (status != EFI_NOT_FOUND);
error = create_efivars_bin_attributes(efivars);
if (error)
unregister_efivars(efivars);
efivars->efi_pstore_info = efi_pstore_info;
efivars->efi_pstore_info.buf = kmalloc(4096, GFP_KERNEL);
if (efivars->efi_pstore_info.buf) {
efivars->efi_pstore_info.bufsize = 1024;
efivars->efi_pstore_info.data = efivars;
spin_lock_init(&efivars->efi_pstore_info.buf_lock);
pstore_register(&efivars->efi_pstore_info);
}
register_filesystem(&efivarfs_type);
out:
kfree(variable_name);
return error;
}
EXPORT_SYMBOL_GPL(register_efivars);
/*
* For now we register the efi subsystem with the firmware subsystem
* and the vars subsystem with the efi subsystem. In the future, it
* might make sense to split off the efi subsystem into its own
* driver, but for now only efivars will register with it, so just
* include it here.
*/
static int __init
efivars_init(void)
{
int error = 0;
printk(KERN_INFO "EFI Variables Facility v%s %s\n", EFIVARS_VERSION,
EFIVARS_DATE);
efi: Make 'efi_enabled' a function to query EFI facilities Originally 'efi_enabled' indicated whether a kernel was booted from EFI firmware. Over time its semantics have changed, and it now indicates whether or not we are booted on an EFI machine with bit-native firmware, e.g. 64-bit kernel with 64-bit firmware. The immediate motivation for this patch is the bug report at, https://bugs.launchpad.net/ubuntu-cdimage/+bug/1040557 which details how running a platform driver on an EFI machine that is designed to run under BIOS can cause the machine to become bricked. Also, the following report, https://bugzilla.kernel.org/show_bug.cgi?id=47121 details how running said driver can also cause Machine Check Exceptions. Drivers need a new means of detecting whether they're running on an EFI machine, as sadly the expression, if (!efi_enabled) hasn't been a sufficient condition for quite some time. Users actually want to query 'efi_enabled' for different reasons - what they really want access to is the list of available EFI facilities. For instance, the x86 reboot code needs to know whether it can invoke the ResetSystem() function provided by the EFI runtime services, while the ACPI OSL code wants to know whether the EFI config tables were mapped successfully. There are also checks in some of the platform driver code to simply see if they're running on an EFI machine (which would make it a bad idea to do BIOS-y things). This patch is a prereq for the samsung-laptop fix patch. Cc: David Airlie <airlied@linux.ie> Cc: Corentin Chary <corentincj@iksaif.net> Cc: Matthew Garrett <mjg59@srcf.ucam.org> Cc: Dave Jiang <dave.jiang@intel.com> Cc: Olof Johansson <olof@lixom.net> Cc: Peter Jones <pjones@redhat.com> Cc: Colin Ian King <colin.king@canonical.com> Cc: Steve Langasek <steve.langasek@canonical.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Konrad Rzeszutek Wilk <konrad@kernel.org> Cc: Rafael J. Wysocki <rjw@sisk.pl> Cc: <stable@vger.kernel.org> Signed-off-by: Matt Fleming <matt.fleming@intel.com> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2012-11-14 09:42:35 +00:00
if (!efi_enabled(EFI_RUNTIME_SERVICES))
return 0;
/* For now we'll register the efi directory at /sys/firmware/efi */
efi_kobj = kobject_create_and_add("efi", firmware_kobj);
if (!efi_kobj) {
printk(KERN_ERR "efivars: Firmware registration failed.\n");
return -ENOMEM;
}
ops.get_variable = efi.get_variable;
ops.set_variable = efi.set_variable;
ops.get_next_variable = efi.get_next_variable;
ops.query_variable_info = efi.query_variable_info;
error = register_efivars(&__efivars, &ops, efi_kobj);
if (error)
goto err_put;
/* Don't forget the systab entry */
error = sysfs_create_group(efi_kobj, &efi_subsys_attr_group);
if (error) {
printk(KERN_ERR
"efivars: Sysfs attribute export failed with error %d.\n",
error);
goto err_unregister;
}
return 0;
err_unregister:
unregister_efivars(&__efivars);
err_put:
kobject_put(efi_kobj);
return error;
}
static void __exit
efivars_exit(void)
{
efi: Make 'efi_enabled' a function to query EFI facilities Originally 'efi_enabled' indicated whether a kernel was booted from EFI firmware. Over time its semantics have changed, and it now indicates whether or not we are booted on an EFI machine with bit-native firmware, e.g. 64-bit kernel with 64-bit firmware. The immediate motivation for this patch is the bug report at, https://bugs.launchpad.net/ubuntu-cdimage/+bug/1040557 which details how running a platform driver on an EFI machine that is designed to run under BIOS can cause the machine to become bricked. Also, the following report, https://bugzilla.kernel.org/show_bug.cgi?id=47121 details how running said driver can also cause Machine Check Exceptions. Drivers need a new means of detecting whether they're running on an EFI machine, as sadly the expression, if (!efi_enabled) hasn't been a sufficient condition for quite some time. Users actually want to query 'efi_enabled' for different reasons - what they really want access to is the list of available EFI facilities. For instance, the x86 reboot code needs to know whether it can invoke the ResetSystem() function provided by the EFI runtime services, while the ACPI OSL code wants to know whether the EFI config tables were mapped successfully. There are also checks in some of the platform driver code to simply see if they're running on an EFI machine (which would make it a bad idea to do BIOS-y things). This patch is a prereq for the samsung-laptop fix patch. Cc: David Airlie <airlied@linux.ie> Cc: Corentin Chary <corentincj@iksaif.net> Cc: Matthew Garrett <mjg59@srcf.ucam.org> Cc: Dave Jiang <dave.jiang@intel.com> Cc: Olof Johansson <olof@lixom.net> Cc: Peter Jones <pjones@redhat.com> Cc: Colin Ian King <colin.king@canonical.com> Cc: Steve Langasek <steve.langasek@canonical.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Konrad Rzeszutek Wilk <konrad@kernel.org> Cc: Rafael J. Wysocki <rjw@sisk.pl> Cc: <stable@vger.kernel.org> Signed-off-by: Matt Fleming <matt.fleming@intel.com> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2012-11-14 09:42:35 +00:00
if (efi_enabled(EFI_RUNTIME_SERVICES)) {
efivars: prevent oops on unload when efi is not enabled efivars_exit() should check for efi_enabled and not undo allocations when efi is not enabled. Otherwise there is an Oops during module unload: calling efivars_init+0x0/0x1000 [efivars] @ 2810 EFI Variables Facility v0.08 2004-May-17 initcall efivars_init+0x0/0x1000 [efivars] returned 0 after 5120 usecs Oops: 0000 [#1] SMP DEBUG_PAGEALLOC last sysfs file: /sys/module/firmware_class/initstate CPU 1 Modules linked in: efivars(-) af_packet tun nfsd lockd nfs_acl auth_rpcgss sunrpc ipt_REJECT nf_conntrack_ipv4 nf_defrag_ipv4 iptable_filter ip_tables ip6t_REJECT xt_tcpudp nf_conntrack_ipv6 nf_defrag_ipv6 xt_state nf_conntrack ip6table_filter ip6_tables x_tables ipv6 cpufreq_ondemand acpi_cpufreq freq_table mperf binfmt_misc dm_mirror dm_region_hash dm_log dm_multipath scsi_dh dm_mod snd_hda_codec_analog snd_hda_intel snd_hda_codec snd_hwdep mousedev snd_seq joydev snd_seq_device mac_hid evdev snd_pcm usbkbd usbmouse usbhid snd_timer hid tg3 snd sr_mod pcspkr rtc_cmos soundcore cdrom iTCO_wdt processor sg dcdbas i2c_i801 rtc_core iTCO_vendor_support intel_agp snd_page_alloc thermal_sys rtc_lib intel_gtt 8250_pnp button hwmon unix ide_pci_generic ide_core ata_generic pata_acpi ata_piix sd_mod crc_t10dif ext3 jbd mbcache uhci_hcd ohci_hcd ssb mmc_core pcmcia pcmcia_core firmware_class ehci_hcd usbcore [last unloaded: dell_rbu] Pid: 2812, comm: rmmod Not tainted 2.6.39-rc6 #1 Dell Inc. OptiPlex 745 /0TY565 RIP: 0010:[<ffffffffa06a17f6>] [<ffffffffa06a17f6>] unregister_efivars+0x28/0x12c [efivars] RSP: 0018:ffff88005eedde98 EFLAGS: 00010283 RAX: ffffffffa06a23fc RBX: ffffffffa06a44c0 RCX: ffff88007c227a50 RDX: 0000000000000000 RSI: 00000055ac13db78 RDI: ffffffffa06a44c0 RBP: ffff88005eeddec8 R08: 0000000000000000 R09: ffff88005eeddd78 R10: ffffffffa06a4220 R11: ffff88005eeddd78 R12: fffffffffffff7d0 R13: 00007fff5a3aaec0 R14: 0000000000000000 R15: ffffffffa06a4508 FS: 00007fa8dcc4a6f0(0000) GS:ffff88007c200000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000000000 CR3: 000000005d148000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process rmmod (pid: 2812, threadinfo ffff88005eedc000, task ffff88006754b000) Stack: ffff88005eeddec8 ffffffffa06a4220 0000000000000000 00007fff5a3aaec0 0000000000000000 0000000000000001 ffff88005eedded8 ffffffffa06a2418 ffff88005eeddf78 ffffffff810d3598 ffffffffa06a4220 0000000000000880 Call Trace: [<ffffffffa06a2418>] efivars_exit+0x1c/0xc04 [efivars] [<ffffffff810d3598>] sys_delete_module+0x2d6/0x368 [<ffffffff812d1db9>] ? lockdep_sys_exit_thunk+0x35/0x67 [<ffffffff810fcba1>] ? audit_syscall_entry+0x172/0x1a5 [<ffffffff81575082>] system_call_fastpath+0x16/0x1b Code: 5c c9 c3 55 48 89 e5 41 57 41 56 41 55 41 54 53 48 83 ec 08 0f 1f 44 00 00 4c 8b 67 48 48 89 fb 4c 8d 7f 48 49 81 ec 30 08 00 00 <4d> 8b ac 24 30 08 00 00 49 81 ed 30 08 00 00 eb 59 48 89 df 48 RIP [<ffffffffa06a17f6>] unregister_efivars+0x28/0x12c [efivars] RSP <ffff88005eedde98> CR2: 0000000000000000 ---[ end trace aa99b99090f70baa ]--- Matt apparently removed such a check in 2004 (with no reason given): * 17 May 2004 - Matt Domsch <Matt_Domsch@dell.com> * remove check for efi_enabled in exit but there have been several changes since then. Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: Mike Waychison <mikew@google.com> Tested-by: Randy Dunlap <randy.dunlap@oracle.com> Cc: Matt Domsch <Matt_Domsch@dell.com> Cc: <matthew.e.tolentino@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2011-05-06 20:27:41 +00:00
unregister_efivars(&__efivars);
kobject_put(efi_kobj);
}
}
module_init(efivars_init);
module_exit(efivars_exit);