btrfs: Fix NULL pointer access in btrfs_check_leaked_roots()
PATCH 1/1 btrfs: Fix NULL pointer access in btrfs_check_leaked_roots()
If fs_info->super_copy or fs_info->super_for_commit is NULL in btrfs_get_tree_subvol(), the btrfs_check_leaked_roots() will get the btrfs_root list entry using the fs_info->allocated_roots->next which is NULL.
syzkaller reported the following information:
------------[ cut here ]------------ BUG: unable to handle page fault for address: fffffffffffffbb0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 64c9067 P4D 64c9067 PUD 64cb067 PMD 0 Oops: Oops: 0000 [#1] SMP KASAN PTI CPU: 0 UID: 0 PID: 1402 Comm: syz.1.35 Not tainted 6.15.8 #4 PREEMPT(lazy) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), (...) RIP: 0010:arch_atomic_read arch/x86/include/asm/atomic.h:23 [inline] RIP: 0010:raw_atomic_read include/linux/atomic/atomic-arch-fallback.h:457 [inline] RIP: 0010:atomic_read include/linux/atomic/atomic-instrumented.h:33 [inline] RIP: 0010:refcount_read include/linux/refcount.h:170 [inline] RIP: 0010:btrfs_check_leaked_roots+0x18f/0x2c0 fs/btrfs/disk-io.c:1230 [...] Call Trace: <TASK> btrfs_free_fs_info+0x310/0x410 fs/btrfs/disk-io.c:1280 btrfs_get_tree_subvol+0x592/0x6b0 fs/btrfs/super.c:2029 btrfs_get_tree+0x63/0x80 fs/btrfs/super.c:2097 vfs_get_tree+0x98/0x320 fs/super.c:1759 do_new_mount+0x357/0x660 fs/namespace.c:3899 path_mount+0x716/0x19c0 fs/namespace.c:4226 do_mount fs/namespace.c:4239 [inline] __do_sys_mount fs/namespace.c:4450 [inline] __se_sys_mount fs/namespace.c:4427 [inline] __x64_sys_mount+0x28c/0x310 fs/namespace.c:4427 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x92/0x180 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f032eaffa8d [...]
This should check if the fs_info->allocated_roots->next is NULL before accessing it.
--- fs/btrfs/disk-io.c | 3 +++ 1 file changed, 3 insertions(+)
fs/btrfs/disk-io.c +3 -0
@@ -1213,6 +1213,9 @@ void btrfs_check_leaked_roots(const struct btrfs_fs_info *fs_info)#ifdef CONFIG_BTRFS_DEBUG struct btrfs_root *root; +if (!fs_info->allocated_roots.next) +return ; +while (!list_empty (&fs_info->allocated_roots)) {char buf[BTRFS_ROOT_NAME_BUF_LEN ];
Signature · 1 line
2.43.5
Quoted 39 lines · Dewei Meng
If
fs_info->super_copyorfs_info->super_for_commitis NULL inbtrfs_get_tree_subvol(), thebtrfs_check_leaked_roots()will get thebtrfs_rootlist entry using thefs_info->allocated_roots->nextwhich is NULL.syzkaller reported the following information:
------------[ cut here ]------------ BUG: unable to handle page fault for address: fffffffffffffbb0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 64c9067 P4D 64c9067 PUD 64cb067 PMD 0 Oops: Oops: 0000 [#1] SMP KASAN PTI CPU: 0 UID: 0 PID: 1402 Comm: syz.1.35 Not tainted 6.15.8 #4 PREEMPT(lazy) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), (...) RIP: 0010:arch_atomic_read arch/x86/include/asm/atomic.h:23 [inline] RIP: 0010:raw_atomic_read include/linux/atomic/atomic-arch-fallback.h:457 [inline] RIP: 0010:atomic_read include/linux/atomic/atomic-instrumented.h:33 [inline] RIP: 0010:refcount_read include/linux/refcount.h:170 [inline] RIP: 0010:btrfs_check_leaked_roots+0x18f/0x2c0 fs/btrfs/disk-io.c:1230 [...] Call Trace: <TASK> btrfs_free_fs_info+0x310/0x410 fs/btrfs/disk-io.c:1280 btrfs_get_tree_subvol+0x592/0x6b0 fs/btrfs/super.c:2029 btrfs_get_tree+0x63/0x80 fs/btrfs/super.c:2097 vfs_get_tree+0x98/0x320 fs/super.c:1759 do_new_mount+0x357/0x660 fs/namespace.c:3899 path_mount+0x716/0x19c0 fs/namespace.c:4226 do_mount fs/namespace.c:4239 [inline] __do_sys_mount fs/namespace.c:4450 [inline] __se_sys_mount fs/namespace.c:4427 [inline] __x64_sys_mount+0x28c/0x310 fs/namespace.c:4427 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x92/0x180 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f032eaffa8d [...]
Quoted 8 lines
This should check if the
fs_info->allocated_roots->nextis NULL before accessing it.Fixes 3bb17a25bcb0 ("btrfs: add get_tree callback for new mount API")Signed-off-by Dewei Meng <[email protected]>--- fs/btrfs/disk-io.c | 3 +++ 1 file changed, 3 insertions(+)
Looks good to me.
Thank you
Quoted 19 lines
fs/btrfs/disk-io.c +3
-1@@ -1213,6 +1213,9 @@ void btrfs_check_leaked_roots(const struct btrfs_fs_info *fs_info)#ifdef CONFIG_BTRFS_DEBUG struct btrfs_root *root; +if (!fs_info->allocated_roots.next) +return ; +while (!list_empty (&fs_info->allocated_roots)) {char buf[BTRFS_ROOT_NAME_BUF_LEN ]; --2.43.5
Quoted 2 lines · Dewei Meng
If
fs_info->super_copyorfs_info->super_for_commitis NULL inbtrfs_get_tree_subvol(),
Please reorganize this sentence. It would be way more easier to read by just saying something like "If memory allocation failed for fs_info->super_copy or fs_info->super_for_commit in btrfs_get_tree_subvol()".
Quoted 51 lines
the
btrfs_check_leaked_roots()will get thebtrfs_rootlist entry using thefs_info->allocated_roots->nextwhich is NULL.syzkaller reported the following information:
------------[ cut here ]------------ BUG: unable to handle page fault for address: fffffffffffffbb0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 64c9067 P4D 64c9067 PUD 64cb067 PMD 0 Oops: Oops: 0000 [#1] SMP KASAN PTI CPU: 0 UID: 0 PID: 1402 Comm: syz.1.35 Not tainted 6.15.8 #4 PREEMPT(lazy) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), (...) RIP: 0010:arch_atomic_read arch/x86/include/asm/atomic.h:23 [inline] RIP: 0010:raw_atomic_read include/linux/atomic/atomic-arch-fallback.h:457 [inline] RIP: 0010:atomic_read include/linux/atomic/atomic-instrumented.h:33 [inline] RIP: 0010:refcount_read include/linux/refcount.h:170 [inline] RIP: 0010:btrfs_check_leaked_roots+0x18f/0x2c0 fs/btrfs/disk-io.c:1230 [...] Call Trace: <TASK> btrfs_free_fs_info+0x310/0x410 fs/btrfs/disk-io.c:1280 btrfs_get_tree_subvol+0x592/0x6b0 fs/btrfs/super.c:2029 btrfs_get_tree+0x63/0x80 fs/btrfs/super.c:2097 vfs_get_tree+0x98/0x320 fs/super.c:1759 do_new_mount+0x357/0x660 fs/namespace.c:3899 path_mount+0x716/0x19c0 fs/namespace.c:4226 do_mount fs/namespace.c:4239 [inline] __do_sys_mount fs/namespace.c:4450 [inline] __se_sys_mount fs/namespace.c:4427 [inline] __x64_sys_mount+0x28c/0x310 fs/namespace.c:4427 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x92/0x180 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f032eaffa8d [...]This should check if the
fs_info->allocated_roots->nextis NULL before accessing it.Fixes 3bb17a25bcb0 ("btrfs: add get_tree callback for new mount API")Signed-off-by Dewei Meng <[email protected]>--- fs/btrfs/disk-io.c | 3 +++ 1 file changed, 3 insertions(+)fs/btrfs/disk-io.c +0
-0
fs/btrfs/disk-io.c void btrfs_check_leaked_roots(const struct btrfs_fs_info *fs_info) · line 1218
Quoted 6 lines
fs/btrfs/disk-io.c +3
-0@@ -1213,6 +1213,9 @@ void btrfs_check_leaked_roots(const struct btrfs_fs_info *fs_info)#ifdef CONFIG_BTRFS_DEBUG struct btrfs_root *root; +if (!fs_info->allocated_roots.next) +return ; +
The check looks too adhoc to me.
It would be much easier to just call kvfree() in the error handling of super_copy/super_for_commit allocation, we do not and should not call btrfs_free_fs_info() before calling btrfs_init_fs_info().
Thanks,
Qu
Quoted 3 lines
fs/btrfs/disk-io.c +0
-0while (!list_empty (&fs_info->allocated_roots)) {char buf[BTRFS_ROOT_NAME_BUF_LEN ];
Quoted 69 lines · Qu Wenruo
Quoted 2 lines · Dewei Meng
If
fs_info->super_copyorfs_info->super_for_commitis NULL inbtrfs_get_tree_subvol(),Please reorganize this sentence. It would be way more easier to read by just saying something like "If memory allocation failed for fs_info-
Quoted 53 lines
super_copyorfs_info->super_for_commitinbtrfs_get_tree_subvol()".the
btrfs_check_leaked_roots()will get thebtrfs_rootlist entry using thefs_info->allocated_roots->nextwhich is NULL.syzkaller reported the following information:
------------[ cut here ]------------ BUG: unable to handle page fault for address: fffffffffffffbb0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 64c9067 P4D 64c9067 PUD 64cb067 PMD 0 Oops: Oops: 0000 [#1] SMP KASAN PTI CPU: 0 UID: 0 PID: 1402 Comm: syz.1.35 Not tainted 6.15.8 #4 PREEMPT(lazy) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), (...) RIP: 0010:arch_atomic_read arch/x86/include/asm/atomic.h:23 [inline] RIP: 0010:raw_atomic_read include/linux/atomic/atomic-arch- fallback.h:457 [inline] RIP: 0010:atomic_read include/linux/atomic/atomic-instrumented.h:33 [inline] RIP: 0010:refcount_read include/linux/refcount.h:170 [inline] RIP: 0010:btrfs_check_leaked_roots+0x18f/0x2c0 fs/btrfs/disk-io.c:1230 [...] Call Trace: <TASK> btrfs_free_fs_info+0x310/0x410 fs/btrfs/disk-io.c:1280 btrfs_get_tree_subvol+0x592/0x6b0 fs/btrfs/super.c:2029 btrfs_get_tree+0x63/0x80 fs/btrfs/super.c:2097 vfs_get_tree+0x98/0x320 fs/super.c:1759 do_new_mount+0x357/0x660 fs/namespace.c:3899 path_mount+0x716/0x19c0 fs/namespace.c:4226 do_mount fs/namespace.c:4239 [inline] __do_sys_mount fs/namespace.c:4450 [inline] __se_sys_mount fs/namespace.c:4427 [inline] __x64_sys_mount+0x28c/0x310 fs/namespace.c:4427 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x92/0x180 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f032eaffa8d[...]
This should check if the
fs_info->allocated_roots->nextis NULL before accessing it.Fixes 3bb17a25bcb0 ("btrfs: add get_tree callback for new mount API")Signed-off-by Dewei Meng <[email protected]>---
fs/btrfs/disk-io.c| 3 +++1 file changed, 3 insertions(+)fs/btrfs/disk-io.c +0
-0fs/btrfs/disk-io.c void btrfs_check_leaked_roots(const struct · line 1218
Quoted 6 lines
fs/btrfs/disk-io.c +3
-0@@ -1213,6 +1213,9 @@ void btrfs_check_leaked_roots(const struct btrfs_fs_info *fs_info) #ifdef CONFIG_BTRFS_DEBUG struct btrfs_root *root; + if (!fs_info->allocated_roots.next) + return; +The check looks too adhoc to me.
Quoted 1 line
It would be much easier to just call
kvfree()
Of course with kfree() on the super_copy/super_for_commit before kvfree() on the fs_info.
Quoted 8 lines
in the error handling of super_copy/super_for_commit allocation, we do not and should not call
btrfs_free_fs_info()before callingbtrfs_init_fs_info().Thanks,
QuQuoted 2 lines
fs/btrfs/disk-io.c +0
-0while (!list_empty(&fs_info->allocated_roots)) { char buf[BTRFS_ROOT_NAME_BUF_LEN];
Quoted 72 lines · Qu Wenruo
Quoted 2 lines · Dewei Meng
If
fs_info->super_copyorfs_info->super_for_commitis NULL inbtrfs_get_tree_subvol(),Please reorganize this sentence. It would be way more easier to read by just saying something like "If memory allocation failed for
fs_info->super_copyorfs_info->super_for_commitinbtrfs_get_tree_subvol()".Quoted 51 lines
the
btrfs_check_leaked_roots()will get thebtrfs_rootlist entry using thefs_info->allocated_roots->nextwhich is NULL.syzkaller reported the following information:
------------[ cut here ]------------ BUG: unable to handle page fault for address: fffffffffffffbb0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 64c9067 P4D 64c9067 PUD 64cb067 PMD 0 Oops: Oops: 0000 [#1] SMP KASAN PTI CPU: 0 UID: 0 PID: 1402 Comm: syz.1.35 Not tainted 6.15.8 #4 PREEMPT(lazy) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), (...) RIP: 0010:arch_atomic_read arch/x86/include/asm/atomic.h:23 [inline] RIP: 0010:raw_atomic_read include/linux/atomic/atomic-arch-fallback.h:457 [inline] RIP: 0010:atomic_read include/linux/atomic/atomic-instrumented.h:33 [inline] RIP: 0010:refcount_read include/linux/refcount.h:170 [inline] RIP: 0010:btrfs_check_leaked_roots+0x18f/0x2c0 fs/btrfs/disk-io.c:1230 [...] Call Trace: <TASK> btrfs_free_fs_info+0x310/0x410 fs/btrfs/disk-io.c:1280 btrfs_get_tree_subvol+0x592/0x6b0 fs/btrfs/super.c:2029 btrfs_get_tree+0x63/0x80 fs/btrfs/super.c:2097 vfs_get_tree+0x98/0x320 fs/super.c:1759 do_new_mount+0x357/0x660 fs/namespace.c:3899 path_mount+0x716/0x19c0 fs/namespace.c:4226 do_mount fs/namespace.c:4239 [inline] __do_sys_mount fs/namespace.c:4450 [inline] __se_sys_mount fs/namespace.c:4427 [inline] __x64_sys_mount+0x28c/0x310 fs/namespace.c:4427 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x92/0x180 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f032eaffa8d [...]This should check if the
fs_info->allocated_roots->nextis NULL before accessing it.Fixes 3bb17a25bcb0 ("btrfs: add get_tree callback for new mount API")Signed-off-by Dewei Meng <[email protected]>--- fs/btrfs/disk-io.c | 3 +++ 1 file changed, 3 insertions(+)fs/btrfs/disk-io.c +0
-0fs/btrfs/disk-io.c void btrfs_check_leaked_roots(const struct btrfs_fs_info *fs_info) · line 1218
Quoted 6 lines
fs/btrfs/disk-io.c +3
-0@@ -1213,6 +1213,9 @@ void btrfs_check_leaked_roots(const struct btrfs_fs_info *fs_info)#ifdef CONFIG_BTRFS_DEBUG struct btrfs_root *root; +if (!fs_info->allocated_roots.next) +return ; +The check looks too adhoc to me.
Quoted 3 lines
It would be much easier to just call
kvfree()in the error handling of super_copy/super_for_commit allocation, we do not and should not callbtrfs_free_fs_info()before callingbtrfs_init_fs_info().
Right. I like this solution better too.
--nX
Quoted 7 lines
Thanks,
QuQuoted 3 lines
fs/btrfs/disk-io.c +0
-0while (!list_empty (&fs_info->allocated_roots)) {char buf[BTRFS_ROOT_NAME_BUF_LEN ];
Quoted 7 lines · Qu Wenruo
Quoted 2 lines · Dewei Meng
If
fs_info->super_copyorfs_info->super_for_commitis NULL inbtrfs_get_tree_subvol(),Please reorganize this sentence. It would be way more easier to read by just saying something like "If memory allocation failed for
fs_info->super_copyorfs_info->super_for_commitinbtrfs_get_tree_subvol()".
I agree, I will fix these words to make them easier to read.
Quoted 61 lines
Quoted 52 lines
the
btrfs_check_leaked_roots()will get thebtrfs_rootlist entry using thefs_info->allocated_roots->nextwhich is NULL.syzkaller reported the following information:
------------[ cut here ]------------ BUG: unable to handle page fault for address: fffffffffffffbb0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 64c9067 P4D 64c9067 PUD 64cb067 PMD 0 Oops: Oops: 0000 [#1] SMP KASAN PTI CPU: 0 UID: 0 PID: 1402 Comm: syz.1.35 Not tainted 6.15.8 #4 PREEMPT(lazy) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), (...) RIP: 0010:arch_atomic_read arch/x86/include/asm/atomic.h:23 [inline] RIP: 0010:raw_atomic_read include/linux/atomic/atomic-arch-fallback.h:457 [inline] RIP: 0010:atomic_read include/linux/atomic/atomic-instrumented.h:33 [inline] RIP: 0010:refcount_read include/linux/refcount.h:170 [inline] RIP: 0010:btrfs_check_leaked_roots+0x18f/0x2c0 fs/btrfs/disk-io.c:1230 [...] Call Trace: <TASK> btrfs_free_fs_info+0x310/0x410 fs/btrfs/disk-io.c:1280 btrfs_get_tree_subvol+0x592/0x6b0 fs/btrfs/super.c:2029 btrfs_get_tree+0x63/0x80 fs/btrfs/super.c:2097 vfs_get_tree+0x98/0x320 fs/super.c:1759 do_new_mount+0x357/0x660 fs/namespace.c:3899 path_mount+0x716/0x19c0 fs/namespace.c:4226 do_mount fs/namespace.c:4239 [inline] __do_sys_mount fs/namespace.c:4450 [inline] __se_sys_mount fs/namespace.c:4427 [inline] __x64_sys_mount+0x28c/0x310 fs/namespace.c:4427 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x92/0x180 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f032eaffa8d[...]
This should check if the
fs_info->allocated_roots->nextis NULL before accessing it.Fixes 3bb17a25bcb0 ("btrfs: add get_tree callback for new mount API")Signed-off-by Dewei Meng <[email protected]>---
fs/btrfs/disk-io.c| 3 +++1 file changed, 3 insertions(+)fs/btrfs/disk-io.c +0
-0fs/btrfs/disk-io.c void btrfs_check_leaked_roots(const struct · line 1218
Quoted 6 lines
fs/btrfs/disk-io.c +2
-0@@ -1213,6 +1213,9 @@ void btrfs_check_leaked_roots(const struct btrfs_fs_info *fs_info) #ifdef CONFIG_BTRFS_DEBUG struct btrfs_root *root; + if (!fs_info->allocated_roots.next) + return; +The check looks too adhoc to me.
Quoted 1 line
It would be much easier to just call
kvfree()in the error handling of super_copy/super_for_commit allocation, we do not and should not callbtrfs_free_fs_info()before callingbtrfs_init_fs_info().
It is a good solution to fix this bug, or can we put the 'btrfs_init_fs_info(fs_info)' before super_copy/super_for_commit allocation?
Thanks,
Dewei Meng
Quoted 8 lines
Thanks,
QuQuoted 2 lines
fs/btrfs/disk-io.c +0
-0while (!list_empty(&fs_info->allocated_roots)) { char buf[BTRFS_ROOT_NAME_BUF_LEN];
Quoted 71 lines · Dewei Meng
Quoted 7 lines · Qu Wenruo
Quoted 2 lines · Dewei Meng
If
fs_info->super_copyorfs_info->super_for_commitis NULL inbtrfs_get_tree_subvol(),Please reorganize this sentence. It would be way more easier to read by just saying something like "If memory allocation failed for
fs_info->super_copyorfs_info->super_for_commitinbtrfs_get_tree_subvol()".I agree, I will fix these words to make them easier to read.
Quoted 61 lines
Quoted 52 lines
the
btrfs_check_leaked_roots()will get thebtrfs_rootlist entry using thefs_info->allocated_roots->nextwhich is NULL.syzkaller reported the following information:
------------[ cut here ]------------ BUG: unable to handle page fault for address: fffffffffffffbb0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 64c9067 P4D 64c9067 PUD 64cb067 PMD 0 Oops: Oops: 0000 [#1] SMP KASAN PTI CPU: 0 UID: 0 PID: 1402 Comm: syz.1.35 Not tainted 6.15.8 #4 PREEMPT(lazy) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), (...) RIP: 0010:arch_atomic_read arch/x86/include/asm/atomic.h:23 [inline] RIP: 0010:raw_atomic_read include/linux/atomic/atomic-arch- fallback.h:457 [inline] RIP: 0010:atomic_read include/linux/atomic/atomic- instrumented.h:33 [inline] RIP: 0010:refcount_read include/linux/refcount.h:170 [inline] RIP: 0010:btrfs_check_leaked_roots+0x18f/0x2c0 fs/btrfs/disk- io.c:1230 [...] Call Trace: <TASK> btrfs_free_fs_info+0x310/0x410 fs/btrfs/disk-io.c:1280 btrfs_get_tree_subvol+0x592/0x6b0 fs/btrfs/super.c:2029 btrfs_get_tree+0x63/0x80 fs/btrfs/super.c:2097 vfs_get_tree+0x98/0x320 fs/super.c:1759 do_new_mount+0x357/0x660 fs/namespace.c:3899 path_mount+0x716/0x19c0 fs/namespace.c:4226 do_mount fs/namespace.c:4239 [inline] __do_sys_mount fs/namespace.c:4450 [inline] __se_sys_mount fs/namespace.c:4427 [inline] __x64_sys_mount+0x28c/0x310 fs/namespace.c:4427 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x92/0x180 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f032eaffa8d[...]
This should check if the
fs_info->allocated_roots->nextis NULL before accessing it.Fixes 3bb17a25bcb0 ("btrfs: add get_tree callback for new mount API")Signed-off-by Dewei Meng <[email protected]>---
fs/btrfs/disk-io.c| 3 +++1 file changed, 3 insertions(+)fs/btrfs/disk-io.c +0
-0fs/btrfs/disk-io.c void btrfs_check_leaked_roots(const struct · line 1218
Quoted 6 lines
fs/btrfs/disk-io.c +2
-0@@ -1213,6 +1213,9 @@ void btrfs_check_leaked_roots(const struct btrfs_fs_info *fs_info) #ifdef CONFIG_BTRFS_DEBUG struct btrfs_root *root; + if (!fs_info->allocated_roots.next) + return; +The check looks too adhoc to me.
Quoted 3 lines
Quoted 1 line
It would be much easier to just call
kvfree()in the error handling of super_copy/super_for_commit allocation, we do not and should not callbtrfs_free_fs_info()before callingbtrfs_init_fs_info().It is a good solution to fix this bug, or can we put the '
btrfs_init_fs_info(fs_info)' before super_copy/super_for_commit allocation?
That also sounds fine to me.
Quoted 13 lines
Thanks,
Dewei Meng
Quoted 8 lines
Thanks,
QuQuoted 2 lines
fs/btrfs/disk-io.c +0
-0while (!list_empty(&fs_info->allocated_roots)) { char buf[BTRFS_ROOT_NAME_BUF_LEN];
Quoted 74 lines · Qu Wenruo
Quoted 71 lines · Dewei Meng
Quoted 7 lines · Qu Wenruo
Quoted 2 lines · Dewei Meng
If
fs_info->super_copyorfs_info->super_for_commitis NULL inbtrfs_get_tree_subvol(),Please reorganize this sentence. It would be way more easier to read by just saying something like "If memory allocation failed for
fs_info->super_copyorfs_info->super_for_commitinbtrfs_get_tree_subvol()".I agree, I will fix these words to make them easier to read.
Quoted 61 lines
Quoted 52 lines
the
btrfs_check_leaked_roots()will get thebtrfs_rootlist entry using thefs_info->allocated_roots->nextwhich is NULL.syzkaller reported the following information:
------------[ cut here ]------------ BUG: unable to handle page fault for address: fffffffffffffbb0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 64c9067 P4D 64c9067 PUD 64cb067 PMD 0 Oops: Oops: 0000 [#1] SMP KASAN PTI CPU: 0 UID: 0 PID: 1402 Comm: syz.1.35 Not tainted 6.15.8 #4 PREEMPT(lazy) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), (...) RIP: 0010:arch_atomic_read arch/x86/include/asm/atomic.h:23 [inline] RIP: 0010:raw_atomic_read include/linux/atomic/atomic-arch- fallback.h:457 [inline] RIP: 0010:atomic_read include/linux/atomic/atomic- instrumented.h:33 [inline] RIP: 0010:refcount_read include/linux/refcount.h:170 [inline] RIP: 0010:btrfs_check_leaked_roots+0x18f/0x2c0 fs/btrfs/disk- io.c:1230 [...] Call Trace: <TASK> btrfs_free_fs_info+0x310/0x410 fs/btrfs/disk-io.c:1280 btrfs_get_tree_subvol+0x592/0x6b0 fs/btrfs/super.c:2029 btrfs_get_tree+0x63/0x80 fs/btrfs/super.c:2097 vfs_get_tree+0x98/0x320 fs/super.c:1759 do_new_mount+0x357/0x660 fs/namespace.c:3899 path_mount+0x716/0x19c0 fs/namespace.c:4226 do_mount fs/namespace.c:4239 [inline] __do_sys_mount fs/namespace.c:4450 [inline] __se_sys_mount fs/namespace.c:4427 [inline] __x64_sys_mount+0x28c/0x310 fs/namespace.c:4427 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x92/0x180 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f032eaffa8d[...]
This should check if the
fs_info->allocated_roots->nextis NULL before accessing it.Fixes 3bb17a25bcb0 ("btrfs: add get_tree callback for new mount API")Signed-off-by Dewei Meng <[email protected]>---
fs/btrfs/disk-io.c| 3 +++1 file changed, 3 insertions(+)fs/btrfs/disk-io.c +0
-0fs/btrfs/disk-io.c void btrfs_check_leaked_roots(const struct · line 1218
Quoted 6 lines
fs/btrfs/disk-io.c +2
-0@@ -1213,6 +1213,9 @@ void btrfs_check_leaked_roots(const struct btrfs_fs_info *fs_info) #ifdef CONFIG_BTRFS_DEBUG struct btrfs_root *root; + if (!fs_info->allocated_roots.next) + return; +The check looks too adhoc to me.
Quoted 5 lines
Quoted 3 lines
Quoted 1 line
It would be much easier to just call
kvfree()in the error handling of super_copy/super_for_commit allocation, we do not and should not callbtrfs_free_fs_info()before callingbtrfs_init_fs_info().It is a good solution to fix this bug, or can we put the '
btrfs_init_fs_info(fs_info)' before super_copy/super_for_commit allocation?That also sounds fine to me.
I have tested the kfree method, and test case runs fine. I plan to take your suggestion and re-do the patch.
Thanks.
Quoted 17 lines
Quoted 14 lines
Thanks,
Dewei Meng
Quoted 8 lines
Thanks,
QuQuoted 2 lines
fs/btrfs/disk-io.c +0
-0while (!list_empty(&fs_info->allocated_roots)) { char buf[BTRFS_ROOT_NAME_BUF_LEN];