efivarfs: expose used and total size
PATCH 1/1 efivarfs: expose used and total size
From: Anisse Astier <[email protected]>
When writing EFI variables, one might get errors with no other message on why it fails. Being able to see how much is used by EFI variables helps analyzing such issues.
Since this is not a conventional filesystem, block size is intentionally set to 1 instead of PAGE_SIZE.
x86 quirks of reserved size are taken into account; so that available and free size can be different, further helping debugging space issues.
With this patch, one can see the remaining space in EFI variable storage via efivarfs, like this:
$ df -h /sys/firmware/efi/efivars/ Filesystem Size Used Avail Use% Mounted on efivarfs 176K 106K 66K 62% /sys/firmware/efi/efivars
[ardb: - rename efi_reserved_space() to efivar_reserved_space()
- whitespace/coding style tweaks]
(cherry-picked from d86ff3333cb1 ("efivarfs: expose used and total size")) Adjusted for headers in linux-6.1.y
---
Cc:Steve McIntyre <[email protected]>
Cc:Richard Hughes <[email protected]>
Background for this backport is that fwupd needs to be able to do CA updates on Debian oldstable (bookworm) which tracks 6.1.7. The CA update process checks for free storage, and needs this function to do it.
arch/x86/platform/efi/quirks.c | 8 +++++++ drivers/firmware/efi/efi.c | 1 + drivers/firmware/efi/vars.c | 12 +++++++++++ fs/efivarfs/super.c | 39 +++++++++++++++++++++++++++++++++- include/linux/efi.h | 11 ++++++++++ 5 files changed, 70 insertions(+), 1 deletion(-)
arch/x86/platform/efi/quirks.c +8 -0
@@ -114,6 +114,14 @@ void efi_delete_dummy_variable(void) EFI_VARIABLE_RUNTIME_ACCESS, 0, NULL); } +u64 efivar_reserved_space (void ) +{ +if (efi_no_storage_paranoia) +return 0 ; +return EFI_MIN_RESERVE ; +} +EXPORT_SYMBOL_GPL (efivar_reserved_space); +/* * In the nonblocking case we do not attempt to perform garbage * collection if we do not have enough free space. Rather, we do the
drivers/firmware/efi/efi.c +1 -0
@@ -211,6 +211,7 @@ static int generic_ops_register(void) generic_ops.get_variable = efi.get_variable; generic_ops.get_next_variable = efi.get_next_variable; generic_ops.query_variable_store = efi_query_variable_store; + generic_ops.query_variable_info = efi.query_variable_info;if (efi_rt_services_supported (EFI_RT_SUPPORTED_SET_VARIABLE )) { generic_ops.set_variable = efi.set_variable;
drivers/firmware/efi/vars.c +12 -0
@@ -250,3 +250,15 @@ efi_status_t efivar_set_variable(efi_char16_t *name, efi_guid_t *vendor,return status; }EXPORT_SYMBOL_NS_GPL (efivar_set_variable,EFIVAR ); + +efi_status_t efivar_query_variable_info (u32 attr, +u64 *storage_space, +u64 *remaining_space, +u64 *max_variable_size) +{ +if (!__efivars->ops->query_variable_info) +return EFI_UNSUPPORTED ; +return __efivars->ops->query_variable_info (attr, storage_space, + remaining_space, max_variable_size); +} +EXPORT_SYMBOL_NS_GPL (efivar_query_variable_info,EFIVAR );
fs/efivarfs/super.c +38 -1
@@ -14,6 +14,7 @@#include <linux/slab.h> #include <linux/magic.h> #include <linux/printk.h> +#include <linux/statfs.h> #include "internal.h"
@@ -24,8 +25,44 @@ static void efivarfs_evict_inode(struct inode *inode)clear_inode (inode); } +static int efivarfs_statfs (struct dentry *dentry,struct kstatfs *buf) +{ +const u32 attr =EFI_VARIABLE_NON_VOLATILE | +EFI_VARIABLE_BOOTSERVICE_ACCESS | +EFI_VARIABLE_RUNTIME_ACCESS ; +u64 storage_space, remaining_space, max_variable_size; +efi_status_t status; + + status =efivar_query_variable_info (attr, &storage_space, &remaining_space, + &max_variable_size); +if (status !=EFI_SUCCESS ) +return efi_status_to_err (status); + +/* +* This is not a normal filesystem, so no point in pretending it has a block +* size; we declare f_bsize to 1, so that we can then report the exact value +* sent by EFI QueryVariableInfo in f_blocks and f_bfree +*/ + buf->f_bsize =1 ; + buf->f_namelen =NAME_MAX ; + buf->f_blocks = storage_space; + buf->f_bfree = remaining_space; + buf->f_type = dentry->d_sb->s_magic; + +/* +* In f_bavail we declare the free space that the kernel will allow writing +* when the storage_paranoia x86 quirk is active. To use more, users +* should boot the kernel with efi_no_storage_paranoia. +*/ +if (remaining_space >efivar_reserved_space ()) + buf->f_bavail = remaining_space -efivar_reserved_space (); +else + buf->f_bavail =0 ; + +return 0 ; +}static const struct super_operations efivarfs_ops = { - .statfs = simple_statfs, + .statfs = efivarfs_statfs, .drop_inode = generic_delete_inode, .evict_inode = efivarfs_evict_inode, };
include/linux/efi.h +11 -0
@@ -1039,6 +1039,7 @@ struct efivar_operations {efi_set_variable_t *set_variable;efi_set_variable_t *set_variable_nonblocking;efi_query_variable_store_t *query_variable_store; +efi_query_variable_info_t *query_variable_info; };struct efivars {
@@ -1047,6 +1048,12 @@ struct efivars {const struct efivar_operations *ops; }; +#ifdef CONFIG_X86 +u64 __attribute_const__ efivar_reserved_space(void); +#else +static inline u64 efivar_reserved_space (void ) {return 0 ; } +#endif +/* * The maximum size of VariableName + Data = 1024 * Therefore, it's reasonable to save that much
@@ -1081,6 +1088,10 @@ efi_status_t efivar_set_variable_locked(efi_char16_t *name, efi_guid_t *vendor,efi_status_t efivar_set_variable (efi_char16_t *name,efi_guid_t *vendor,u32 attr,unsigned long data_size,void *data); +efi_status_t efivar_query_variable_info (u32 attr,u64 *storage_space, +u64 *remaining_space, +u64 *max_variable_size); +#if IS_ENABLED (CONFIG_EFI_CAPSULE_LOADER )extern bool efi_capsule_pending (int *reset_type);
Signature · 1 line
2.53.0
Quoted 25 lines · Mario Limonciello (AMD)
From: Anisse Astier <[email protected]>
When writing EFI variables, one might get errors with no other message on why it fails. Being able to see how much is used by EFI variables helps analyzing such issues.
Since this is not a conventional filesystem, block size is intentionally set to 1 instead of
PAGE_SIZE.x86 quirks of reserved size are taken into account; so that available and free size can be different, further helping debugging space issues.
With this patch, one can see the remaining space in EFI variable storage via efivarfs, like this:
$ df -h /sys/firmware/efi/efivars/ Filesystem Size Used Avail Use% Mounted on efivarfs 176K 106K 66K 62% /sys/firmware/efi/efivarsSigned-off-by Anisse Astier <[email protected]>[ardb: - rename
efi_reserved_space()toefivar_reserved_space()
- whitespace/coding style tweaks]Signed-off-by Ard Biesheuvel <[email protected]>(cherry-picked from
d86ff3333cb1("efivarfs: expose used and total size")) Adjusted for headers in linux-6.1.y
Quoted 8 lines
Signed-off-by Mario Limonciello <[email protected]>---
Cc:Steve McIntyre <[email protected]>
Cc:Richard Hughes <[email protected]>Background for this backport is that fwupd needs to be able to do CA updates on Debian oldstable (bookworm) which tracks 6.1.7. The CA update process checks for free storage, and needs this function to do it.
What is the git id of this commit?
Quoted 31 lines · Greg KH
Quoted 30 lines · Mario Limonciello (AMD)
From: Anisse Astier <[email protected]>
When writing EFI variables, one might get errors with no other message on why it fails. Being able to see how much is used by EFI variables helps analyzing such issues.
Since this is not a conventional filesystem, block size is intentionally set to 1 instead of
PAGE_SIZE.x86 quirks of reserved size are taken into account; so that available and free size can be different, further helping debugging space issues.
With this patch, one can see the remaining space in EFI variable storage via efivarfs, like this:
$ df -h /sys/firmware/efi/efivars/ Filesystem Size Used Avail Use% Mounted on efivarfs 176K 106K 66K 62% /sys/firmware/efi/efivarsSigned-off-by Anisse Astier <[email protected]>[ardb: - rename
efi_reserved_space()toefivar_reserved_space()
- whitespace/coding style tweaks]Signed-off-by Ard Biesheuvel <[email protected]>(cherry-picked from
d86ff3333cb1("efivarfs: expose used and total size")) Adjusted for headers in linux-6.1.ySigned-off-by Mario Limonciello <[email protected]>---
Cc:Steve McIntyre <[email protected]>
Cc:Richard Hughes <[email protected]>
Quoted 5 lines
Quoted 3 lines
Background for this backport is that fwupd needs to be able to do CA updates on Debian oldstable (bookworm) which tracks 6.1.7. The CA update process checks for free storage, and needs this function to do it.
What is the git id of this commit?
Oh I mentioned it above between Ard's S-o-b and mine:
d86ff3333cb1 ("efivarfs: expose used and total size")
Quoted 38 lines · Mario Limonciello
Quoted 35 lines · Greg KH
Quoted 33 lines · Mario Limonciello (AMD)
From: Anisse Astier <[email protected]>
When writing EFI variables, one might get errors with no other message on why it fails. Being able to see how much is used by EFI variables helps analyzing such issues.
Since this is not a conventional filesystem, block size is intentionally set to 1 instead of
PAGE_SIZE.x86 quirks of reserved size are taken into account; so that available and free size can be different, further helping debugging space issues.
With this patch, one can see the remaining space in EFI variable storage via efivarfs, like this:
$ df -h /sys/firmware/efi/efivars/ Filesystem Size Used Avail Use% Mounted on efivarfs 176K 106K 66K 62% /sys/firmware/efi/efivarsSigned-off-by Anisse Astier <[email protected]>[ardb: - rename
efi_reserved_space()toefivar_reserved_space()
- whitespace/coding style tweaks]Signed-off-by Ard Biesheuvel <[email protected]>(cherry-picked from
d86ff3333cb1("efivarfs: expose used and total size")) Adjusted for headers in linux-6.1.ySigned-off-by Mario Limonciello <[email protected]>---
Cc:Steve McIntyre <[email protected]>
Cc:Richard Hughes <[email protected]>Background for this backport is that fwupd needs to be able to do CA updates on Debian oldstable (bookworm) which tracks 6.1.7. The CA update process checks for free storage, and needs this function to do it.
Quoted 5 lines
Quoted 1 line
What is the git id of this commit?
Oh I mentioned it above between Ard's S-o-b and mine:
d86ff3333cb1("efivarfs: expose used and total size")
ick, you did, sorry, I missed it as that's not the "normal" cherry-pick output, my fault
Sasha Levin
maintainerQuoted 3 lines
Background for this backport is that fwupd needs to be able to do CA updates on Debian oldstable (bookworm) which tracks 6.1.7. The CA update process checks for free storage, and needs this function to do it.
Queued for 6.1. I also pulled in the follow-up fix 79b83606abc7 ("efivarfs: fix statfs() on efivarfs") on top, so the broken statfs() error path this commit shipped with doesn't land in 6.1.y, thanks.
Signature · 2 lines
Thanks, Sasha
Quoted 8 lines · Sasha Levin
Quoted 3 lines
Background for this backport is that fwupd needs to be able to do CA updates on Debian oldstable (bookworm) which tracks 6.1.7. The CA update process checks for free storage, and needs this function to do it.
Queued for 6.1. I also pulled in the follow-up fix
79b83606abc7("efivarfs: fixstatfs()on efivarfs") on top, so the brokenstatfs()error path this commit shipped with doesn't land in 6.1.y, thanks.
Much appreciated.