2013-02-08 23:48:51 +08:00
|
|
|
#include <linux/efi.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/pstore.h>
|
2013-05-02 08:51:54 +08:00
|
|
|
#include <linux/slab.h>
|
2013-04-30 18:30:24 +08:00
|
|
|
#include <linux/ucs2_string.h>
|
2013-02-08 23:48:51 +08:00
|
|
|
|
|
|
|
#define DUMP_NAME_LEN 52
|
|
|
|
|
|
|
|
static bool efivars_pstore_disable =
|
|
|
|
IS_ENABLED(CONFIG_EFI_VARS_PSTORE_DEFAULT_DISABLE);
|
|
|
|
|
|
|
|
module_param_named(pstore_disable, efivars_pstore_disable, bool, 0644);
|
|
|
|
|
|
|
|
#define PSTORE_EFI_ATTRIBUTES \
|
|
|
|
(EFI_VARIABLE_NON_VOLATILE | \
|
|
|
|
EFI_VARIABLE_BOOTSERVICE_ACCESS | \
|
|
|
|
EFI_VARIABLE_RUNTIME_ACCESS)
|
|
|
|
|
|
|
|
static int efi_pstore_open(struct pstore_info *psi)
|
|
|
|
{
|
|
|
|
psi->data = NULL;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int efi_pstore_close(struct pstore_info *psi)
|
|
|
|
{
|
|
|
|
psi->data = NULL;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
struct pstore_read_data {
|
|
|
|
u64 *id;
|
|
|
|
enum pstore_type_id *type;
|
|
|
|
int *count;
|
|
|
|
struct timespec *timespec;
|
2013-08-17 04:57:51 +08:00
|
|
|
bool *compressed;
|
2013-02-08 23:48:51 +08:00
|
|
|
char **buf;
|
|
|
|
};
|
|
|
|
|
2013-11-29 15:58:57 +08:00
|
|
|
static inline u64 generic_id(unsigned long timestamp,
|
|
|
|
unsigned int part, int count)
|
|
|
|
{
|
2014-06-09 22:50:40 +08:00
|
|
|
return ((u64) timestamp * 100 + part) * 1000 + count;
|
2013-11-29 15:58:57 +08:00
|
|
|
}
|
|
|
|
|
2013-02-08 23:48:51 +08:00
|
|
|
static int efi_pstore_read_func(struct efivar_entry *entry, void *data)
|
|
|
|
{
|
|
|
|
efi_guid_t vendor = LINUX_EFI_CRASH_GUID;
|
|
|
|
struct pstore_read_data *cb_data = data;
|
2013-08-17 04:57:51 +08:00
|
|
|
char name[DUMP_NAME_LEN], data_type;
|
2013-02-08 23:48:51 +08:00
|
|
|
int i;
|
|
|
|
int cnt;
|
|
|
|
unsigned int part;
|
|
|
|
unsigned long time, size;
|
|
|
|
|
|
|
|
if (efi_guidcmp(entry->var.VendorGuid, vendor))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
for (i = 0; i < DUMP_NAME_LEN; i++)
|
|
|
|
name[i] = entry->var.VariableName[i];
|
|
|
|
|
2013-08-17 04:57:51 +08:00
|
|
|
if (sscanf(name, "dump-type%u-%u-%d-%lu-%c",
|
|
|
|
cb_data->type, &part, &cnt, &time, &data_type) == 5) {
|
2013-11-29 15:58:57 +08:00
|
|
|
*cb_data->id = generic_id(time, part, cnt);
|
2013-08-17 04:57:51 +08:00
|
|
|
*cb_data->count = cnt;
|
|
|
|
cb_data->timespec->tv_sec = time;
|
|
|
|
cb_data->timespec->tv_nsec = 0;
|
|
|
|
if (data_type == 'C')
|
|
|
|
*cb_data->compressed = true;
|
|
|
|
else
|
|
|
|
*cb_data->compressed = false;
|
|
|
|
} else if (sscanf(name, "dump-type%u-%u-%d-%lu",
|
2013-02-08 23:48:51 +08:00
|
|
|
cb_data->type, &part, &cnt, &time) == 4) {
|
2013-11-29 15:58:57 +08:00
|
|
|
*cb_data->id = generic_id(time, part, cnt);
|
2013-02-08 23:48:51 +08:00
|
|
|
*cb_data->count = cnt;
|
|
|
|
cb_data->timespec->tv_sec = time;
|
|
|
|
cb_data->timespec->tv_nsec = 0;
|
2013-08-17 04:57:51 +08:00
|
|
|
*cb_data->compressed = false;
|
2013-02-08 23:48:51 +08:00
|
|
|
} else if (sscanf(name, "dump-type%u-%u-%lu",
|
|
|
|
cb_data->type, &part, &time) == 3) {
|
|
|
|
/*
|
|
|
|
* Check if an old format,
|
|
|
|
* which doesn't support holding
|
|
|
|
* multiple logs, remains.
|
|
|
|
*/
|
2013-11-29 15:58:57 +08:00
|
|
|
*cb_data->id = generic_id(time, part, 0);
|
2013-02-08 23:48:51 +08:00
|
|
|
*cb_data->count = 0;
|
|
|
|
cb_data->timespec->tv_sec = time;
|
|
|
|
cb_data->timespec->tv_nsec = 0;
|
2013-08-17 04:57:51 +08:00
|
|
|
*cb_data->compressed = false;
|
2013-02-08 23:48:51 +08:00
|
|
|
} else
|
|
|
|
return 0;
|
|
|
|
|
2013-04-30 03:08:02 +08:00
|
|
|
entry->var.DataSize = 1024;
|
|
|
|
__efivar_entry_get(entry, &entry->var.Attributes,
|
|
|
|
&entry->var.DataSize, entry->var.Data);
|
|
|
|
size = entry->var.DataSize;
|
efivars, efi-pstore: Hold off deletion of sysfs entry until the scan is completed
Currently, when mounting pstore file system, a read callback of
efi_pstore driver runs mutiple times as below.
- In the first read callback, scan efivar_sysfs_list from head and pass
a kmsg buffer of a entry to an upper pstore layer.
- In the second read callback, rescan efivar_sysfs_list from the entry
and pass another kmsg buffer to it.
- Repeat the scan and pass until the end of efivar_sysfs_list.
In this process, an entry is read across the multiple read function
calls. To avoid race between the read and erasion, the whole process
above is protected by a spinlock, holding in open() and releasing in
close().
At the same time, kmemdup() is called to pass the buffer to pstore
filesystem during it. And then, it causes a following lockdep warning.
To make the dynamic memory allocation runnable without taking spinlock,
holding off a deletion of sysfs entry if it happens while scanning it
via efi_pstore, and deleting it after the scan is completed.
To implement it, this patch introduces two flags, scanning and deleting,
to efivar_entry.
On the code basis, it seems that all the scanning and deleting logic is
not needed because __efivars->lock are not dropped when reading from the
EFI variable store.
But, the scanning and deleting logic is still needed because an
efi-pstore and a pstore filesystem works as follows.
In case an entry(A) is found, the pointer is saved to psi->data. And
efi_pstore_read() passes the entry(A) to a pstore filesystem by
releasing __efivars->lock.
And then, the pstore filesystem calls efi_pstore_read() again and the
same entry(A), which is saved to psi->data, is used for resuming to scan
a sysfs-list.
So, to protect the entry(A), the logic is needed.
[ 1.143710] ------------[ cut here ]------------
[ 1.144058] WARNING: CPU: 1 PID: 1 at kernel/lockdep.c:2740 lockdep_trace_alloc+0x104/0x110()
[ 1.144058] DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags))
[ 1.144058] Modules linked in:
[ 1.144058] CPU: 1 PID: 1 Comm: systemd Not tainted 3.11.0-rc5 #2
[ 1.144058] 0000000000000009 ffff8800797e9ae0 ffffffff816614a5 ffff8800797e9b28
[ 1.144058] ffff8800797e9b18 ffffffff8105510d 0000000000000080 0000000000000046
[ 1.144058] 00000000000000d0 00000000000003af ffffffff81ccd0c0 ffff8800797e9b78
[ 1.144058] Call Trace:
[ 1.144058] [<ffffffff816614a5>] dump_stack+0x54/0x74
[ 1.144058] [<ffffffff8105510d>] warn_slowpath_common+0x7d/0xa0
[ 1.144058] [<ffffffff8105517c>] warn_slowpath_fmt+0x4c/0x50
[ 1.144058] [<ffffffff8131290f>] ? vsscanf+0x57f/0x7b0
[ 1.144058] [<ffffffff810bbd74>] lockdep_trace_alloc+0x104/0x110
[ 1.144058] [<ffffffff81192da0>] __kmalloc_track_caller+0x50/0x280
[ 1.144058] [<ffffffff815147bb>] ? efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff8115b260>] kmemdup+0x20/0x50
[ 1.144058] [<ffffffff815147bb>] efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff81514800>] ? efi_pstore_read_func.part.1+0x170/0x170
[ 1.144058] [<ffffffff815148b4>] efi_pstore_read_func+0xb4/0xe0
[ 1.144058] [<ffffffff81512b7b>] __efivar_entry_iter+0xfb/0x120
[ 1.144058] [<ffffffff8151428f>] efi_pstore_read+0x3f/0x50
[ 1.144058] [<ffffffff8128d7ba>] pstore_get_records+0x9a/0x150
[ 1.158207] [<ffffffff812af25c>] ? selinux_d_instantiate+0x1c/0x20
[ 1.158207] [<ffffffff8128ce30>] ? parse_options+0x80/0x80
[ 1.158207] [<ffffffff8128ced5>] pstore_fill_super+0xa5/0xc0
[ 1.158207] [<ffffffff811ae7d2>] mount_single+0xa2/0xd0
[ 1.158207] [<ffffffff8128ccf8>] pstore_mount+0x18/0x20
[ 1.158207] [<ffffffff811ae8b9>] mount_fs+0x39/0x1b0
[ 1.158207] [<ffffffff81160550>] ? __alloc_percpu+0x10/0x20
[ 1.158207] [<ffffffff811c9493>] vfs_kern_mount+0x63/0xf0
[ 1.158207] [<ffffffff811cbb0e>] do_mount+0x23e/0xa20
[ 1.158207] [<ffffffff8115b51b>] ? strndup_user+0x4b/0xf0
[ 1.158207] [<ffffffff811cc373>] SyS_mount+0x83/0xc0
[ 1.158207] [<ffffffff81673cc2>] system_call_fastpath+0x16/0x1b
[ 1.158207] ---[ end trace 61981bc62de9f6f4 ]---
Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com>
Tested-by: Madper Xie <cxie@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
2013-10-31 03:27:26 +08:00
|
|
|
memcpy(*cb_data->buf, entry->var.Data,
|
|
|
|
(size_t)min_t(unsigned long, EFIVARS_DATA_SIZE_MAX, size));
|
2013-04-30 03:08:02 +08:00
|
|
|
|
2013-02-08 23:48:51 +08:00
|
|
|
return size;
|
|
|
|
}
|
|
|
|
|
efivars, efi-pstore: Hold off deletion of sysfs entry until the scan is completed
Currently, when mounting pstore file system, a read callback of
efi_pstore driver runs mutiple times as below.
- In the first read callback, scan efivar_sysfs_list from head and pass
a kmsg buffer of a entry to an upper pstore layer.
- In the second read callback, rescan efivar_sysfs_list from the entry
and pass another kmsg buffer to it.
- Repeat the scan and pass until the end of efivar_sysfs_list.
In this process, an entry is read across the multiple read function
calls. To avoid race between the read and erasion, the whole process
above is protected by a spinlock, holding in open() and releasing in
close().
At the same time, kmemdup() is called to pass the buffer to pstore
filesystem during it. And then, it causes a following lockdep warning.
To make the dynamic memory allocation runnable without taking spinlock,
holding off a deletion of sysfs entry if it happens while scanning it
via efi_pstore, and deleting it after the scan is completed.
To implement it, this patch introduces two flags, scanning and deleting,
to efivar_entry.
On the code basis, it seems that all the scanning and deleting logic is
not needed because __efivars->lock are not dropped when reading from the
EFI variable store.
But, the scanning and deleting logic is still needed because an
efi-pstore and a pstore filesystem works as follows.
In case an entry(A) is found, the pointer is saved to psi->data. And
efi_pstore_read() passes the entry(A) to a pstore filesystem by
releasing __efivars->lock.
And then, the pstore filesystem calls efi_pstore_read() again and the
same entry(A), which is saved to psi->data, is used for resuming to scan
a sysfs-list.
So, to protect the entry(A), the logic is needed.
[ 1.143710] ------------[ cut here ]------------
[ 1.144058] WARNING: CPU: 1 PID: 1 at kernel/lockdep.c:2740 lockdep_trace_alloc+0x104/0x110()
[ 1.144058] DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags))
[ 1.144058] Modules linked in:
[ 1.144058] CPU: 1 PID: 1 Comm: systemd Not tainted 3.11.0-rc5 #2
[ 1.144058] 0000000000000009 ffff8800797e9ae0 ffffffff816614a5 ffff8800797e9b28
[ 1.144058] ffff8800797e9b18 ffffffff8105510d 0000000000000080 0000000000000046
[ 1.144058] 00000000000000d0 00000000000003af ffffffff81ccd0c0 ffff8800797e9b78
[ 1.144058] Call Trace:
[ 1.144058] [<ffffffff816614a5>] dump_stack+0x54/0x74
[ 1.144058] [<ffffffff8105510d>] warn_slowpath_common+0x7d/0xa0
[ 1.144058] [<ffffffff8105517c>] warn_slowpath_fmt+0x4c/0x50
[ 1.144058] [<ffffffff8131290f>] ? vsscanf+0x57f/0x7b0
[ 1.144058] [<ffffffff810bbd74>] lockdep_trace_alloc+0x104/0x110
[ 1.144058] [<ffffffff81192da0>] __kmalloc_track_caller+0x50/0x280
[ 1.144058] [<ffffffff815147bb>] ? efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff8115b260>] kmemdup+0x20/0x50
[ 1.144058] [<ffffffff815147bb>] efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff81514800>] ? efi_pstore_read_func.part.1+0x170/0x170
[ 1.144058] [<ffffffff815148b4>] efi_pstore_read_func+0xb4/0xe0
[ 1.144058] [<ffffffff81512b7b>] __efivar_entry_iter+0xfb/0x120
[ 1.144058] [<ffffffff8151428f>] efi_pstore_read+0x3f/0x50
[ 1.144058] [<ffffffff8128d7ba>] pstore_get_records+0x9a/0x150
[ 1.158207] [<ffffffff812af25c>] ? selinux_d_instantiate+0x1c/0x20
[ 1.158207] [<ffffffff8128ce30>] ? parse_options+0x80/0x80
[ 1.158207] [<ffffffff8128ced5>] pstore_fill_super+0xa5/0xc0
[ 1.158207] [<ffffffff811ae7d2>] mount_single+0xa2/0xd0
[ 1.158207] [<ffffffff8128ccf8>] pstore_mount+0x18/0x20
[ 1.158207] [<ffffffff811ae8b9>] mount_fs+0x39/0x1b0
[ 1.158207] [<ffffffff81160550>] ? __alloc_percpu+0x10/0x20
[ 1.158207] [<ffffffff811c9493>] vfs_kern_mount+0x63/0xf0
[ 1.158207] [<ffffffff811cbb0e>] do_mount+0x23e/0xa20
[ 1.158207] [<ffffffff8115b51b>] ? strndup_user+0x4b/0xf0
[ 1.158207] [<ffffffff811cc373>] SyS_mount+0x83/0xc0
[ 1.158207] [<ffffffff81673cc2>] system_call_fastpath+0x16/0x1b
[ 1.158207] ---[ end trace 61981bc62de9f6f4 ]---
Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com>
Tested-by: Madper Xie <cxie@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
2013-10-31 03:27:26 +08:00
|
|
|
/**
|
|
|
|
* efi_pstore_scan_sysfs_enter
|
|
|
|
* @entry: scanning entry
|
|
|
|
* @next: next entry
|
|
|
|
* @head: list head
|
|
|
|
*/
|
|
|
|
static void efi_pstore_scan_sysfs_enter(struct efivar_entry *pos,
|
|
|
|
struct efivar_entry *next,
|
|
|
|
struct list_head *head)
|
|
|
|
{
|
|
|
|
pos->scanning = true;
|
|
|
|
if (&next->list != head)
|
|
|
|
next->scanning = true;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* __efi_pstore_scan_sysfs_exit
|
|
|
|
* @entry: deleting entry
|
|
|
|
* @turn_off_scanning: Check if a scanning flag should be turned off
|
|
|
|
*/
|
|
|
|
static inline void __efi_pstore_scan_sysfs_exit(struct efivar_entry *entry,
|
|
|
|
bool turn_off_scanning)
|
|
|
|
{
|
|
|
|
if (entry->deleting) {
|
|
|
|
list_del(&entry->list);
|
|
|
|
efivar_entry_iter_end();
|
|
|
|
efivar_unregister(entry);
|
|
|
|
efivar_entry_iter_begin();
|
|
|
|
} else if (turn_off_scanning)
|
|
|
|
entry->scanning = false;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* efi_pstore_scan_sysfs_exit
|
|
|
|
* @pos: scanning entry
|
|
|
|
* @next: next entry
|
|
|
|
* @head: list head
|
|
|
|
* @stop: a flag checking if scanning will stop
|
|
|
|
*/
|
|
|
|
static void efi_pstore_scan_sysfs_exit(struct efivar_entry *pos,
|
|
|
|
struct efivar_entry *next,
|
|
|
|
struct list_head *head, bool stop)
|
|
|
|
{
|
|
|
|
__efi_pstore_scan_sysfs_exit(pos, true);
|
|
|
|
if (stop)
|
|
|
|
__efi_pstore_scan_sysfs_exit(next, &next->list != head);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* efi_pstore_sysfs_entry_iter
|
|
|
|
*
|
|
|
|
* @data: function-specific data to pass to callback
|
|
|
|
* @pos: entry to begin iterating from
|
|
|
|
*
|
|
|
|
* You MUST call efivar_enter_iter_begin() before this function, and
|
|
|
|
* efivar_entry_iter_end() afterwards.
|
|
|
|
*
|
|
|
|
* It is possible to begin iteration from an arbitrary entry within
|
|
|
|
* the list by passing @pos. @pos is updated on return to point to
|
|
|
|
* the next entry of the last one passed to efi_pstore_read_func().
|
|
|
|
* To begin iterating from the beginning of the list @pos must be %NULL.
|
|
|
|
*/
|
|
|
|
static int efi_pstore_sysfs_entry_iter(void *data, struct efivar_entry **pos)
|
|
|
|
{
|
|
|
|
struct efivar_entry *entry, *n;
|
|
|
|
struct list_head *head = &efivar_sysfs_list;
|
|
|
|
int size = 0;
|
|
|
|
|
|
|
|
if (!*pos) {
|
|
|
|
list_for_each_entry_safe(entry, n, head, list) {
|
|
|
|
efi_pstore_scan_sysfs_enter(entry, n, head);
|
|
|
|
|
|
|
|
size = efi_pstore_read_func(entry, data);
|
|
|
|
efi_pstore_scan_sysfs_exit(entry, n, head, size < 0);
|
|
|
|
if (size)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
*pos = n;
|
|
|
|
return size;
|
|
|
|
}
|
|
|
|
|
|
|
|
list_for_each_entry_safe_from((*pos), n, head, list) {
|
|
|
|
efi_pstore_scan_sysfs_enter((*pos), n, head);
|
|
|
|
|
|
|
|
size = efi_pstore_read_func((*pos), data);
|
|
|
|
efi_pstore_scan_sysfs_exit((*pos), n, head, size < 0);
|
|
|
|
if (size)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
*pos = n;
|
|
|
|
return size;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* efi_pstore_read
|
|
|
|
*
|
|
|
|
* This function returns a size of NVRAM entry logged via efi_pstore_write().
|
|
|
|
* The meaning and behavior of efi_pstore/pstore are as below.
|
|
|
|
*
|
|
|
|
* size > 0: Got data of an entry logged via efi_pstore_write() successfully,
|
|
|
|
* and pstore filesystem will continue reading subsequent entries.
|
|
|
|
* size == 0: Entry was not logged via efi_pstore_write(),
|
|
|
|
* and efi_pstore driver will continue reading subsequent entries.
|
|
|
|
* size < 0: Failed to get data of entry logging via efi_pstore_write(),
|
|
|
|
* and pstore will stop reading entry.
|
|
|
|
*/
|
2013-02-08 23:48:51 +08:00
|
|
|
static ssize_t efi_pstore_read(u64 *id, enum pstore_type_id *type,
|
|
|
|
int *count, struct timespec *timespec,
|
2013-08-17 04:53:19 +08:00
|
|
|
char **buf, bool *compressed,
|
|
|
|
struct pstore_info *psi)
|
2013-02-08 23:48:51 +08:00
|
|
|
{
|
|
|
|
struct pstore_read_data data;
|
efivars, efi-pstore: Hold off deletion of sysfs entry until the scan is completed
Currently, when mounting pstore file system, a read callback of
efi_pstore driver runs mutiple times as below.
- In the first read callback, scan efivar_sysfs_list from head and pass
a kmsg buffer of a entry to an upper pstore layer.
- In the second read callback, rescan efivar_sysfs_list from the entry
and pass another kmsg buffer to it.
- Repeat the scan and pass until the end of efivar_sysfs_list.
In this process, an entry is read across the multiple read function
calls. To avoid race between the read and erasion, the whole process
above is protected by a spinlock, holding in open() and releasing in
close().
At the same time, kmemdup() is called to pass the buffer to pstore
filesystem during it. And then, it causes a following lockdep warning.
To make the dynamic memory allocation runnable without taking spinlock,
holding off a deletion of sysfs entry if it happens while scanning it
via efi_pstore, and deleting it after the scan is completed.
To implement it, this patch introduces two flags, scanning and deleting,
to efivar_entry.
On the code basis, it seems that all the scanning and deleting logic is
not needed because __efivars->lock are not dropped when reading from the
EFI variable store.
But, the scanning and deleting logic is still needed because an
efi-pstore and a pstore filesystem works as follows.
In case an entry(A) is found, the pointer is saved to psi->data. And
efi_pstore_read() passes the entry(A) to a pstore filesystem by
releasing __efivars->lock.
And then, the pstore filesystem calls efi_pstore_read() again and the
same entry(A), which is saved to psi->data, is used for resuming to scan
a sysfs-list.
So, to protect the entry(A), the logic is needed.
[ 1.143710] ------------[ cut here ]------------
[ 1.144058] WARNING: CPU: 1 PID: 1 at kernel/lockdep.c:2740 lockdep_trace_alloc+0x104/0x110()
[ 1.144058] DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags))
[ 1.144058] Modules linked in:
[ 1.144058] CPU: 1 PID: 1 Comm: systemd Not tainted 3.11.0-rc5 #2
[ 1.144058] 0000000000000009 ffff8800797e9ae0 ffffffff816614a5 ffff8800797e9b28
[ 1.144058] ffff8800797e9b18 ffffffff8105510d 0000000000000080 0000000000000046
[ 1.144058] 00000000000000d0 00000000000003af ffffffff81ccd0c0 ffff8800797e9b78
[ 1.144058] Call Trace:
[ 1.144058] [<ffffffff816614a5>] dump_stack+0x54/0x74
[ 1.144058] [<ffffffff8105510d>] warn_slowpath_common+0x7d/0xa0
[ 1.144058] [<ffffffff8105517c>] warn_slowpath_fmt+0x4c/0x50
[ 1.144058] [<ffffffff8131290f>] ? vsscanf+0x57f/0x7b0
[ 1.144058] [<ffffffff810bbd74>] lockdep_trace_alloc+0x104/0x110
[ 1.144058] [<ffffffff81192da0>] __kmalloc_track_caller+0x50/0x280
[ 1.144058] [<ffffffff815147bb>] ? efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff8115b260>] kmemdup+0x20/0x50
[ 1.144058] [<ffffffff815147bb>] efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff81514800>] ? efi_pstore_read_func.part.1+0x170/0x170
[ 1.144058] [<ffffffff815148b4>] efi_pstore_read_func+0xb4/0xe0
[ 1.144058] [<ffffffff81512b7b>] __efivar_entry_iter+0xfb/0x120
[ 1.144058] [<ffffffff8151428f>] efi_pstore_read+0x3f/0x50
[ 1.144058] [<ffffffff8128d7ba>] pstore_get_records+0x9a/0x150
[ 1.158207] [<ffffffff812af25c>] ? selinux_d_instantiate+0x1c/0x20
[ 1.158207] [<ffffffff8128ce30>] ? parse_options+0x80/0x80
[ 1.158207] [<ffffffff8128ced5>] pstore_fill_super+0xa5/0xc0
[ 1.158207] [<ffffffff811ae7d2>] mount_single+0xa2/0xd0
[ 1.158207] [<ffffffff8128ccf8>] pstore_mount+0x18/0x20
[ 1.158207] [<ffffffff811ae8b9>] mount_fs+0x39/0x1b0
[ 1.158207] [<ffffffff81160550>] ? __alloc_percpu+0x10/0x20
[ 1.158207] [<ffffffff811c9493>] vfs_kern_mount+0x63/0xf0
[ 1.158207] [<ffffffff811cbb0e>] do_mount+0x23e/0xa20
[ 1.158207] [<ffffffff8115b51b>] ? strndup_user+0x4b/0xf0
[ 1.158207] [<ffffffff811cc373>] SyS_mount+0x83/0xc0
[ 1.158207] [<ffffffff81673cc2>] system_call_fastpath+0x16/0x1b
[ 1.158207] ---[ end trace 61981bc62de9f6f4 ]---
Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com>
Tested-by: Madper Xie <cxie@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
2013-10-31 03:27:26 +08:00
|
|
|
ssize_t size;
|
2013-02-08 23:48:51 +08:00
|
|
|
|
|
|
|
data.id = id;
|
|
|
|
data.type = type;
|
|
|
|
data.count = count;
|
|
|
|
data.timespec = timespec;
|
2013-08-17 04:57:51 +08:00
|
|
|
data.compressed = compressed;
|
2013-02-08 23:48:51 +08:00
|
|
|
data.buf = buf;
|
|
|
|
|
efivars, efi-pstore: Hold off deletion of sysfs entry until the scan is completed
Currently, when mounting pstore file system, a read callback of
efi_pstore driver runs mutiple times as below.
- In the first read callback, scan efivar_sysfs_list from head and pass
a kmsg buffer of a entry to an upper pstore layer.
- In the second read callback, rescan efivar_sysfs_list from the entry
and pass another kmsg buffer to it.
- Repeat the scan and pass until the end of efivar_sysfs_list.
In this process, an entry is read across the multiple read function
calls. To avoid race between the read and erasion, the whole process
above is protected by a spinlock, holding in open() and releasing in
close().
At the same time, kmemdup() is called to pass the buffer to pstore
filesystem during it. And then, it causes a following lockdep warning.
To make the dynamic memory allocation runnable without taking spinlock,
holding off a deletion of sysfs entry if it happens while scanning it
via efi_pstore, and deleting it after the scan is completed.
To implement it, this patch introduces two flags, scanning and deleting,
to efivar_entry.
On the code basis, it seems that all the scanning and deleting logic is
not needed because __efivars->lock are not dropped when reading from the
EFI variable store.
But, the scanning and deleting logic is still needed because an
efi-pstore and a pstore filesystem works as follows.
In case an entry(A) is found, the pointer is saved to psi->data. And
efi_pstore_read() passes the entry(A) to a pstore filesystem by
releasing __efivars->lock.
And then, the pstore filesystem calls efi_pstore_read() again and the
same entry(A), which is saved to psi->data, is used for resuming to scan
a sysfs-list.
So, to protect the entry(A), the logic is needed.
[ 1.143710] ------------[ cut here ]------------
[ 1.144058] WARNING: CPU: 1 PID: 1 at kernel/lockdep.c:2740 lockdep_trace_alloc+0x104/0x110()
[ 1.144058] DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags))
[ 1.144058] Modules linked in:
[ 1.144058] CPU: 1 PID: 1 Comm: systemd Not tainted 3.11.0-rc5 #2
[ 1.144058] 0000000000000009 ffff8800797e9ae0 ffffffff816614a5 ffff8800797e9b28
[ 1.144058] ffff8800797e9b18 ffffffff8105510d 0000000000000080 0000000000000046
[ 1.144058] 00000000000000d0 00000000000003af ffffffff81ccd0c0 ffff8800797e9b78
[ 1.144058] Call Trace:
[ 1.144058] [<ffffffff816614a5>] dump_stack+0x54/0x74
[ 1.144058] [<ffffffff8105510d>] warn_slowpath_common+0x7d/0xa0
[ 1.144058] [<ffffffff8105517c>] warn_slowpath_fmt+0x4c/0x50
[ 1.144058] [<ffffffff8131290f>] ? vsscanf+0x57f/0x7b0
[ 1.144058] [<ffffffff810bbd74>] lockdep_trace_alloc+0x104/0x110
[ 1.144058] [<ffffffff81192da0>] __kmalloc_track_caller+0x50/0x280
[ 1.144058] [<ffffffff815147bb>] ? efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff8115b260>] kmemdup+0x20/0x50
[ 1.144058] [<ffffffff815147bb>] efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff81514800>] ? efi_pstore_read_func.part.1+0x170/0x170
[ 1.144058] [<ffffffff815148b4>] efi_pstore_read_func+0xb4/0xe0
[ 1.144058] [<ffffffff81512b7b>] __efivar_entry_iter+0xfb/0x120
[ 1.144058] [<ffffffff8151428f>] efi_pstore_read+0x3f/0x50
[ 1.144058] [<ffffffff8128d7ba>] pstore_get_records+0x9a/0x150
[ 1.158207] [<ffffffff812af25c>] ? selinux_d_instantiate+0x1c/0x20
[ 1.158207] [<ffffffff8128ce30>] ? parse_options+0x80/0x80
[ 1.158207] [<ffffffff8128ced5>] pstore_fill_super+0xa5/0xc0
[ 1.158207] [<ffffffff811ae7d2>] mount_single+0xa2/0xd0
[ 1.158207] [<ffffffff8128ccf8>] pstore_mount+0x18/0x20
[ 1.158207] [<ffffffff811ae8b9>] mount_fs+0x39/0x1b0
[ 1.158207] [<ffffffff81160550>] ? __alloc_percpu+0x10/0x20
[ 1.158207] [<ffffffff811c9493>] vfs_kern_mount+0x63/0xf0
[ 1.158207] [<ffffffff811cbb0e>] do_mount+0x23e/0xa20
[ 1.158207] [<ffffffff8115b51b>] ? strndup_user+0x4b/0xf0
[ 1.158207] [<ffffffff811cc373>] SyS_mount+0x83/0xc0
[ 1.158207] [<ffffffff81673cc2>] system_call_fastpath+0x16/0x1b
[ 1.158207] ---[ end trace 61981bc62de9f6f4 ]---
Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com>
Tested-by: Madper Xie <cxie@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
2013-10-31 03:27:26 +08:00
|
|
|
*data.buf = kzalloc(EFIVARS_DATA_SIZE_MAX, GFP_KERNEL);
|
|
|
|
if (!*data.buf)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
efivar_entry_iter_begin();
|
|
|
|
size = efi_pstore_sysfs_entry_iter(&data,
|
|
|
|
(struct efivar_entry **)&psi->data);
|
|
|
|
efivar_entry_iter_end();
|
|
|
|
if (size <= 0)
|
|
|
|
kfree(*data.buf);
|
|
|
|
return size;
|
2013-02-08 23:48:51 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static int efi_pstore_write(enum pstore_type_id type,
|
|
|
|
enum kmsg_dump_reason reason, u64 *id,
|
2013-08-17 04:52:47 +08:00
|
|
|
unsigned int part, int count, bool compressed, size_t size,
|
2013-02-08 23:48:51 +08:00
|
|
|
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;
|
|
|
|
int i, ret = 0;
|
|
|
|
|
2013-08-17 04:57:51 +08:00
|
|
|
sprintf(name, "dump-type%u-%u-%d-%lu-%c", type, part, count,
|
|
|
|
get_seconds(), compressed ? 'C' : 'D');
|
2013-02-08 23:48:51 +08:00
|
|
|
|
|
|
|
for (i = 0; i < DUMP_NAME_LEN; i++)
|
|
|
|
efi_name[i] = name[i];
|
|
|
|
|
|
|
|
efivar_entry_set_safe(efi_name, vendor, PSTORE_EFI_ATTRIBUTES,
|
|
|
|
!pstore_cannot_block_path(reason),
|
|
|
|
size, psi->buf);
|
|
|
|
|
|
|
|
if (reason == KMSG_DUMP_OOPS)
|
|
|
|
efivar_run_worker();
|
|
|
|
|
|
|
|
*id = part;
|
|
|
|
return ret;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct pstore_erase_data {
|
|
|
|
u64 id;
|
|
|
|
enum pstore_type_id type;
|
|
|
|
int count;
|
|
|
|
struct timespec time;
|
|
|
|
efi_char16_t *name;
|
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Clean up an entry with the same name
|
|
|
|
*/
|
|
|
|
static int efi_pstore_erase_func(struct efivar_entry *entry, void *data)
|
|
|
|
{
|
|
|
|
struct pstore_erase_data *ed = data;
|
|
|
|
efi_guid_t vendor = LINUX_EFI_CRASH_GUID;
|
|
|
|
efi_char16_t efi_name_old[DUMP_NAME_LEN];
|
|
|
|
efi_char16_t *efi_name = ed->name;
|
2013-04-30 18:30:24 +08:00
|
|
|
unsigned long ucs2_len = ucs2_strlen(ed->name);
|
2013-02-08 23:48:51 +08:00
|
|
|
char name_old[DUMP_NAME_LEN];
|
|
|
|
int i;
|
|
|
|
|
|
|
|
if (efi_guidcmp(entry->var.VendorGuid, vendor))
|
|
|
|
return 0;
|
|
|
|
|
2013-04-30 18:30:24 +08:00
|
|
|
if (ucs2_strncmp(entry->var.VariableName,
|
|
|
|
efi_name, (size_t)ucs2_len)) {
|
2013-02-08 23:48:51 +08:00
|
|
|
/*
|
|
|
|
* Check if an old format, which doesn't support
|
|
|
|
* holding multiple logs, remains.
|
|
|
|
*/
|
|
|
|
sprintf(name_old, "dump-type%u-%u-%lu", ed->type,
|
|
|
|
(unsigned int)ed->id, ed->time.tv_sec);
|
|
|
|
|
|
|
|
for (i = 0; i < DUMP_NAME_LEN; i++)
|
|
|
|
efi_name_old[i] = name_old[i];
|
|
|
|
|
2013-04-30 18:30:24 +08:00
|
|
|
if (ucs2_strncmp(entry->var.VariableName, efi_name_old,
|
|
|
|
ucs2_strlen(efi_name_old)))
|
2013-02-08 23:48:51 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
efivars, efi-pstore: Hold off deletion of sysfs entry until the scan is completed
Currently, when mounting pstore file system, a read callback of
efi_pstore driver runs mutiple times as below.
- In the first read callback, scan efivar_sysfs_list from head and pass
a kmsg buffer of a entry to an upper pstore layer.
- In the second read callback, rescan efivar_sysfs_list from the entry
and pass another kmsg buffer to it.
- Repeat the scan and pass until the end of efivar_sysfs_list.
In this process, an entry is read across the multiple read function
calls. To avoid race between the read and erasion, the whole process
above is protected by a spinlock, holding in open() and releasing in
close().
At the same time, kmemdup() is called to pass the buffer to pstore
filesystem during it. And then, it causes a following lockdep warning.
To make the dynamic memory allocation runnable without taking spinlock,
holding off a deletion of sysfs entry if it happens while scanning it
via efi_pstore, and deleting it after the scan is completed.
To implement it, this patch introduces two flags, scanning and deleting,
to efivar_entry.
On the code basis, it seems that all the scanning and deleting logic is
not needed because __efivars->lock are not dropped when reading from the
EFI variable store.
But, the scanning and deleting logic is still needed because an
efi-pstore and a pstore filesystem works as follows.
In case an entry(A) is found, the pointer is saved to psi->data. And
efi_pstore_read() passes the entry(A) to a pstore filesystem by
releasing __efivars->lock.
And then, the pstore filesystem calls efi_pstore_read() again and the
same entry(A), which is saved to psi->data, is used for resuming to scan
a sysfs-list.
So, to protect the entry(A), the logic is needed.
[ 1.143710] ------------[ cut here ]------------
[ 1.144058] WARNING: CPU: 1 PID: 1 at kernel/lockdep.c:2740 lockdep_trace_alloc+0x104/0x110()
[ 1.144058] DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags))
[ 1.144058] Modules linked in:
[ 1.144058] CPU: 1 PID: 1 Comm: systemd Not tainted 3.11.0-rc5 #2
[ 1.144058] 0000000000000009 ffff8800797e9ae0 ffffffff816614a5 ffff8800797e9b28
[ 1.144058] ffff8800797e9b18 ffffffff8105510d 0000000000000080 0000000000000046
[ 1.144058] 00000000000000d0 00000000000003af ffffffff81ccd0c0 ffff8800797e9b78
[ 1.144058] Call Trace:
[ 1.144058] [<ffffffff816614a5>] dump_stack+0x54/0x74
[ 1.144058] [<ffffffff8105510d>] warn_slowpath_common+0x7d/0xa0
[ 1.144058] [<ffffffff8105517c>] warn_slowpath_fmt+0x4c/0x50
[ 1.144058] [<ffffffff8131290f>] ? vsscanf+0x57f/0x7b0
[ 1.144058] [<ffffffff810bbd74>] lockdep_trace_alloc+0x104/0x110
[ 1.144058] [<ffffffff81192da0>] __kmalloc_track_caller+0x50/0x280
[ 1.144058] [<ffffffff815147bb>] ? efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff8115b260>] kmemdup+0x20/0x50
[ 1.144058] [<ffffffff815147bb>] efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff81514800>] ? efi_pstore_read_func.part.1+0x170/0x170
[ 1.144058] [<ffffffff815148b4>] efi_pstore_read_func+0xb4/0xe0
[ 1.144058] [<ffffffff81512b7b>] __efivar_entry_iter+0xfb/0x120
[ 1.144058] [<ffffffff8151428f>] efi_pstore_read+0x3f/0x50
[ 1.144058] [<ffffffff8128d7ba>] pstore_get_records+0x9a/0x150
[ 1.158207] [<ffffffff812af25c>] ? selinux_d_instantiate+0x1c/0x20
[ 1.158207] [<ffffffff8128ce30>] ? parse_options+0x80/0x80
[ 1.158207] [<ffffffff8128ced5>] pstore_fill_super+0xa5/0xc0
[ 1.158207] [<ffffffff811ae7d2>] mount_single+0xa2/0xd0
[ 1.158207] [<ffffffff8128ccf8>] pstore_mount+0x18/0x20
[ 1.158207] [<ffffffff811ae8b9>] mount_fs+0x39/0x1b0
[ 1.158207] [<ffffffff81160550>] ? __alloc_percpu+0x10/0x20
[ 1.158207] [<ffffffff811c9493>] vfs_kern_mount+0x63/0xf0
[ 1.158207] [<ffffffff811cbb0e>] do_mount+0x23e/0xa20
[ 1.158207] [<ffffffff8115b51b>] ? strndup_user+0x4b/0xf0
[ 1.158207] [<ffffffff811cc373>] SyS_mount+0x83/0xc0
[ 1.158207] [<ffffffff81673cc2>] system_call_fastpath+0x16/0x1b
[ 1.158207] ---[ end trace 61981bc62de9f6f4 ]---
Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com>
Tested-by: Madper Xie <cxie@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
2013-10-31 03:27:26 +08:00
|
|
|
if (entry->scanning) {
|
|
|
|
/*
|
|
|
|
* Skip deletion because this entry will be deleted
|
|
|
|
* after scanning is completed.
|
|
|
|
*/
|
|
|
|
entry->deleting = true;
|
|
|
|
} else
|
|
|
|
list_del(&entry->list);
|
|
|
|
|
2013-02-08 23:48:51 +08:00
|
|
|
/* found */
|
|
|
|
__efivar_entry_delete(entry);
|
2013-04-30 03:06:37 +08:00
|
|
|
|
2013-02-08 23:48:51 +08:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int efi_pstore_erase(enum pstore_type_id type, u64 id, int count,
|
|
|
|
struct timespec time, struct pstore_info *psi)
|
|
|
|
{
|
|
|
|
struct pstore_erase_data edata;
|
2013-04-30 02:31:45 +08:00
|
|
|
struct efivar_entry *entry = NULL;
|
2013-02-08 23:48:51 +08:00
|
|
|
char name[DUMP_NAME_LEN];
|
|
|
|
efi_char16_t efi_name[DUMP_NAME_LEN];
|
|
|
|
int found, i;
|
2013-11-29 15:58:57 +08:00
|
|
|
unsigned int part;
|
2013-02-08 23:48:51 +08:00
|
|
|
|
2013-11-29 15:58:57 +08:00
|
|
|
do_div(id, 1000);
|
|
|
|
part = do_div(id, 100);
|
|
|
|
sprintf(name, "dump-type%u-%u-%d-%lu", type, part, count, time.tv_sec);
|
2013-02-08 23:48:51 +08:00
|
|
|
|
|
|
|
for (i = 0; i < DUMP_NAME_LEN; i++)
|
|
|
|
efi_name[i] = name[i];
|
|
|
|
|
2013-11-29 15:58:57 +08:00
|
|
|
edata.id = part;
|
2013-02-08 23:48:51 +08:00
|
|
|
edata.type = type;
|
|
|
|
edata.count = count;
|
|
|
|
edata.time = time;
|
|
|
|
edata.name = efi_name;
|
|
|
|
|
|
|
|
efivar_entry_iter_begin();
|
|
|
|
found = __efivar_entry_iter(efi_pstore_erase_func, &efivar_sysfs_list, &edata, &entry);
|
|
|
|
|
efivars, efi-pstore: Hold off deletion of sysfs entry until the scan is completed
Currently, when mounting pstore file system, a read callback of
efi_pstore driver runs mutiple times as below.
- In the first read callback, scan efivar_sysfs_list from head and pass
a kmsg buffer of a entry to an upper pstore layer.
- In the second read callback, rescan efivar_sysfs_list from the entry
and pass another kmsg buffer to it.
- Repeat the scan and pass until the end of efivar_sysfs_list.
In this process, an entry is read across the multiple read function
calls. To avoid race between the read and erasion, the whole process
above is protected by a spinlock, holding in open() and releasing in
close().
At the same time, kmemdup() is called to pass the buffer to pstore
filesystem during it. And then, it causes a following lockdep warning.
To make the dynamic memory allocation runnable without taking spinlock,
holding off a deletion of sysfs entry if it happens while scanning it
via efi_pstore, and deleting it after the scan is completed.
To implement it, this patch introduces two flags, scanning and deleting,
to efivar_entry.
On the code basis, it seems that all the scanning and deleting logic is
not needed because __efivars->lock are not dropped when reading from the
EFI variable store.
But, the scanning and deleting logic is still needed because an
efi-pstore and a pstore filesystem works as follows.
In case an entry(A) is found, the pointer is saved to psi->data. And
efi_pstore_read() passes the entry(A) to a pstore filesystem by
releasing __efivars->lock.
And then, the pstore filesystem calls efi_pstore_read() again and the
same entry(A), which is saved to psi->data, is used for resuming to scan
a sysfs-list.
So, to protect the entry(A), the logic is needed.
[ 1.143710] ------------[ cut here ]------------
[ 1.144058] WARNING: CPU: 1 PID: 1 at kernel/lockdep.c:2740 lockdep_trace_alloc+0x104/0x110()
[ 1.144058] DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags))
[ 1.144058] Modules linked in:
[ 1.144058] CPU: 1 PID: 1 Comm: systemd Not tainted 3.11.0-rc5 #2
[ 1.144058] 0000000000000009 ffff8800797e9ae0 ffffffff816614a5 ffff8800797e9b28
[ 1.144058] ffff8800797e9b18 ffffffff8105510d 0000000000000080 0000000000000046
[ 1.144058] 00000000000000d0 00000000000003af ffffffff81ccd0c0 ffff8800797e9b78
[ 1.144058] Call Trace:
[ 1.144058] [<ffffffff816614a5>] dump_stack+0x54/0x74
[ 1.144058] [<ffffffff8105510d>] warn_slowpath_common+0x7d/0xa0
[ 1.144058] [<ffffffff8105517c>] warn_slowpath_fmt+0x4c/0x50
[ 1.144058] [<ffffffff8131290f>] ? vsscanf+0x57f/0x7b0
[ 1.144058] [<ffffffff810bbd74>] lockdep_trace_alloc+0x104/0x110
[ 1.144058] [<ffffffff81192da0>] __kmalloc_track_caller+0x50/0x280
[ 1.144058] [<ffffffff815147bb>] ? efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff8115b260>] kmemdup+0x20/0x50
[ 1.144058] [<ffffffff815147bb>] efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff81514800>] ? efi_pstore_read_func.part.1+0x170/0x170
[ 1.144058] [<ffffffff815148b4>] efi_pstore_read_func+0xb4/0xe0
[ 1.144058] [<ffffffff81512b7b>] __efivar_entry_iter+0xfb/0x120
[ 1.144058] [<ffffffff8151428f>] efi_pstore_read+0x3f/0x50
[ 1.144058] [<ffffffff8128d7ba>] pstore_get_records+0x9a/0x150
[ 1.158207] [<ffffffff812af25c>] ? selinux_d_instantiate+0x1c/0x20
[ 1.158207] [<ffffffff8128ce30>] ? parse_options+0x80/0x80
[ 1.158207] [<ffffffff8128ced5>] pstore_fill_super+0xa5/0xc0
[ 1.158207] [<ffffffff811ae7d2>] mount_single+0xa2/0xd0
[ 1.158207] [<ffffffff8128ccf8>] pstore_mount+0x18/0x20
[ 1.158207] [<ffffffff811ae8b9>] mount_fs+0x39/0x1b0
[ 1.158207] [<ffffffff81160550>] ? __alloc_percpu+0x10/0x20
[ 1.158207] [<ffffffff811c9493>] vfs_kern_mount+0x63/0xf0
[ 1.158207] [<ffffffff811cbb0e>] do_mount+0x23e/0xa20
[ 1.158207] [<ffffffff8115b51b>] ? strndup_user+0x4b/0xf0
[ 1.158207] [<ffffffff811cc373>] SyS_mount+0x83/0xc0
[ 1.158207] [<ffffffff81673cc2>] system_call_fastpath+0x16/0x1b
[ 1.158207] ---[ end trace 61981bc62de9f6f4 ]---
Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com>
Tested-by: Madper Xie <cxie@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
2013-10-31 03:27:26 +08:00
|
|
|
if (found && !entry->scanning) {
|
|
|
|
efivar_entry_iter_end();
|
2013-02-08 23:48:51 +08:00
|
|
|
efivar_unregister(entry);
|
efivars, efi-pstore: Hold off deletion of sysfs entry until the scan is completed
Currently, when mounting pstore file system, a read callback of
efi_pstore driver runs mutiple times as below.
- In the first read callback, scan efivar_sysfs_list from head and pass
a kmsg buffer of a entry to an upper pstore layer.
- In the second read callback, rescan efivar_sysfs_list from the entry
and pass another kmsg buffer to it.
- Repeat the scan and pass until the end of efivar_sysfs_list.
In this process, an entry is read across the multiple read function
calls. To avoid race between the read and erasion, the whole process
above is protected by a spinlock, holding in open() and releasing in
close().
At the same time, kmemdup() is called to pass the buffer to pstore
filesystem during it. And then, it causes a following lockdep warning.
To make the dynamic memory allocation runnable without taking spinlock,
holding off a deletion of sysfs entry if it happens while scanning it
via efi_pstore, and deleting it after the scan is completed.
To implement it, this patch introduces two flags, scanning and deleting,
to efivar_entry.
On the code basis, it seems that all the scanning and deleting logic is
not needed because __efivars->lock are not dropped when reading from the
EFI variable store.
But, the scanning and deleting logic is still needed because an
efi-pstore and a pstore filesystem works as follows.
In case an entry(A) is found, the pointer is saved to psi->data. And
efi_pstore_read() passes the entry(A) to a pstore filesystem by
releasing __efivars->lock.
And then, the pstore filesystem calls efi_pstore_read() again and the
same entry(A), which is saved to psi->data, is used for resuming to scan
a sysfs-list.
So, to protect the entry(A), the logic is needed.
[ 1.143710] ------------[ cut here ]------------
[ 1.144058] WARNING: CPU: 1 PID: 1 at kernel/lockdep.c:2740 lockdep_trace_alloc+0x104/0x110()
[ 1.144058] DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags))
[ 1.144058] Modules linked in:
[ 1.144058] CPU: 1 PID: 1 Comm: systemd Not tainted 3.11.0-rc5 #2
[ 1.144058] 0000000000000009 ffff8800797e9ae0 ffffffff816614a5 ffff8800797e9b28
[ 1.144058] ffff8800797e9b18 ffffffff8105510d 0000000000000080 0000000000000046
[ 1.144058] 00000000000000d0 00000000000003af ffffffff81ccd0c0 ffff8800797e9b78
[ 1.144058] Call Trace:
[ 1.144058] [<ffffffff816614a5>] dump_stack+0x54/0x74
[ 1.144058] [<ffffffff8105510d>] warn_slowpath_common+0x7d/0xa0
[ 1.144058] [<ffffffff8105517c>] warn_slowpath_fmt+0x4c/0x50
[ 1.144058] [<ffffffff8131290f>] ? vsscanf+0x57f/0x7b0
[ 1.144058] [<ffffffff810bbd74>] lockdep_trace_alloc+0x104/0x110
[ 1.144058] [<ffffffff81192da0>] __kmalloc_track_caller+0x50/0x280
[ 1.144058] [<ffffffff815147bb>] ? efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff8115b260>] kmemdup+0x20/0x50
[ 1.144058] [<ffffffff815147bb>] efi_pstore_read_func.part.1+0x12b/0x170
[ 1.144058] [<ffffffff81514800>] ? efi_pstore_read_func.part.1+0x170/0x170
[ 1.144058] [<ffffffff815148b4>] efi_pstore_read_func+0xb4/0xe0
[ 1.144058] [<ffffffff81512b7b>] __efivar_entry_iter+0xfb/0x120
[ 1.144058] [<ffffffff8151428f>] efi_pstore_read+0x3f/0x50
[ 1.144058] [<ffffffff8128d7ba>] pstore_get_records+0x9a/0x150
[ 1.158207] [<ffffffff812af25c>] ? selinux_d_instantiate+0x1c/0x20
[ 1.158207] [<ffffffff8128ce30>] ? parse_options+0x80/0x80
[ 1.158207] [<ffffffff8128ced5>] pstore_fill_super+0xa5/0xc0
[ 1.158207] [<ffffffff811ae7d2>] mount_single+0xa2/0xd0
[ 1.158207] [<ffffffff8128ccf8>] pstore_mount+0x18/0x20
[ 1.158207] [<ffffffff811ae8b9>] mount_fs+0x39/0x1b0
[ 1.158207] [<ffffffff81160550>] ? __alloc_percpu+0x10/0x20
[ 1.158207] [<ffffffff811c9493>] vfs_kern_mount+0x63/0xf0
[ 1.158207] [<ffffffff811cbb0e>] do_mount+0x23e/0xa20
[ 1.158207] [<ffffffff8115b51b>] ? strndup_user+0x4b/0xf0
[ 1.158207] [<ffffffff811cc373>] SyS_mount+0x83/0xc0
[ 1.158207] [<ffffffff81673cc2>] system_call_fastpath+0x16/0x1b
[ 1.158207] ---[ end trace 61981bc62de9f6f4 ]---
Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com>
Tested-by: Madper Xie <cxie@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
2013-10-31 03:27:26 +08:00
|
|
|
} else
|
|
|
|
efivar_entry_iter_end();
|
2013-02-08 23:48:51 +08:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct pstore_info efi_pstore_info = {
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.name = "efi",
|
2013-12-19 07:17:10 +08:00
|
|
|
.flags = PSTORE_FLAGS_FRAGILE,
|
2013-02-08 23:48:51 +08:00
|
|
|
.open = efi_pstore_open,
|
|
|
|
.close = efi_pstore_close,
|
|
|
|
.read = efi_pstore_read,
|
|
|
|
.write = efi_pstore_write,
|
|
|
|
.erase = efi_pstore_erase,
|
|
|
|
};
|
|
|
|
|
|
|
|
static __init int efivars_pstore_init(void)
|
|
|
|
{
|
|
|
|
if (!efi_enabled(EFI_RUNTIME_SERVICES))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (!efivars_kobject())
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (efivars_pstore_disable)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
efi_pstore_info.buf = kmalloc(4096, GFP_KERNEL);
|
|
|
|
if (!efi_pstore_info.buf)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
efi_pstore_info.bufsize = 1024;
|
|
|
|
spin_lock_init(&efi_pstore_info.buf_lock);
|
|
|
|
|
2013-06-29 04:14:11 +08:00
|
|
|
if (pstore_register(&efi_pstore_info)) {
|
|
|
|
kfree(efi_pstore_info.buf);
|
|
|
|
efi_pstore_info.buf = NULL;
|
|
|
|
efi_pstore_info.bufsize = 0;
|
|
|
|
}
|
2013-02-08 23:48:51 +08:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static __exit void efivars_pstore_exit(void)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
module_init(efivars_pstore_init);
|
|
|
|
module_exit(efivars_pstore_exit);
|
|
|
|
|
|
|
|
MODULE_DESCRIPTION("EFI variable backend for pstore");
|
|
|
|
MODULE_LICENSE("GPL");
|