Azat Khuzhin
df5314f27d
Use current_time() over CURRENT_TIME macro
...
And this wasn't done undef #ifdef due to we already had current_time()
users
2018-04-20 02:30:19 +03:00
Barry Song
90e4702d37
fix ‘struct file’ has no member named ‘f_dentry’
...
Removed-in: torvalds/linux@78d28e651f
("kill f_dentry macro")
Signed-off-by: Barry Song <21cnbao@gmail.com>
Closes : #10 (picked)
2018-04-20 02:30:19 +03:00
Sankar சங்கர்
af55617de2
Merge pull request #9 from gshrikant/master
...
Fix usage of CURRENT_TIME macro.
2017-06-14 10:18:18 +05:30
Shrikant Giridhar
38669257c5
Fix usage of CURRENT_TIME macro.
...
The CURRENT_TIME macro is no longer valid with the latest kernel
(commit bfe1c566453a0979c0b3cd3728d0de962272f034). Replace usage
with current_time() function.
2017-06-14 00:38:19 -04:00
Azat Khuzhin
5d00eebd45
Merge branch 'journal-v6'
...
This patchset adds journal support into simplefs using jbd2, the only
available journal is external (you could create it using mke2fs -O
journal_dev, for more info look into simple-test.sh), and the only spot
that have start/write_access/dirty_metadata/stop is simplefs_write()
(IOW no support for creation).
This branch is compacted, for more verbose versions of this patchset
please look into more older version (v4, v3, ...).
Here is the example of records in journal:
==> final-jbd2-ext4.logdump <==
logdump
Journal starts at block 1, transaction 4
Found expected sequence 4, type 1 (descriptor block) at block 1
Found expected sequence 4, type 2 (commit block) at block 16
Found expected sequence 5, type 1 (descriptor block) at block 17
Found expected sequence 5, type 2 (commit block) at block 19
Found expected sequence 6, type 1 (descriptor block) at block 20
Found expected sequence 6, type 2 (commit block) at block 22
No magic number at block 23: end of journal.
==> final-jbd2-simplefs.logdump <==
debugfs: logdump -f /dev/loop0
Ext2 superblock header found.
Journal starts at block 2, transaction 2
Found expected sequence 2, type 1 (descriptor block) at block 2
Found expected sequence 2, type 2 (commit block) at block 4
Found expected sequence 3, type 1 (descriptor block) at block 5
Found expected sequence 3, type 2 (commit block) at block 7
Found expected sequence 4, type 1 (descriptor block) at block 8
Found expected sequence 4, type 2 (commit block) at block 10
Found expected sequence 5, type 1 (descriptor block) at block 11
Found expected sequence 5, type 2 (commit block) at block 13
Found expected sequence 6, type 1 (descriptor block) at block 14
Found expected sequence 6, type 2 (commit block) at block 16
Found expected sequence 7, type 1 (descriptor block) at block 17
Found expected sequence 7, type 2 (commit block) at block 19
Found expected sequence 8, type 1 (descriptor block) at block 20
Found expected sequence 8, type 2 (commit block) at block 22
No magic number at block 23: end of journal.
* journal-v6:
simplefs_write: install h_sync (sync on close) for jbd2 handle
Use max available journal size with jbd2_journal_init_dev()
simple-test: create journal with 4k block size
simple-test: use losetup+journal_path=/dev/loopX for mounting with journal
Handle bdevs for journal_path= mount flag
Actually load journal by calling jbd2_journal_load()
simple-test: mount fs with journal
Implement journal_path option
Add loader for internal journal (we need to update mkfs to support this)
Add sfs_trace for lookup and compare inside it
simple-test: enable jbd2-debug
Check that journal was initialized successfully
Write journal device name
Implement external journal device instead of internal
simplefs_write: add journaling using jbd2
Initial support of journaling
2014-10-02 02:27:23 +04:00
Azat Khuzhin
4c486f751b
simplefs_write: install h_sync (sync on close) for jbd2 handle
...
This will make simplefs journaling more robust.
+ head -n100 /proc/fs/jbd2/loop0/info
7 transactions (7 requested), each up to 640 blocks
average:
0ms waiting for transaction
10ms request delay
26ms running transaction
0ms transaction was being locked
0ms flushing data (in ordered mode)
0ms logging transaction
3459us average transaction commit time
1 handles per transaction
1 blocks per transaction
2 logged blocks per transaction
2014-10-02 02:20:37 +04:00
Azat Khuzhin
49937e8acf
Use max available journal size with jbd2_journal_init_dev()
...
Avoid next error:
JBD2: journal file too short
2014-10-02 02:20:37 +04:00
Azat Khuzhin
7023aebf18
simple-test: create journal with 4k block size
...
This must avoid next error:
[ 6.694339] Journal device is: unknown-block(7,0)
[ 6.695620] JBD2: no valid journal superblock found
2014-10-02 02:20:37 +04:00
Azat Khuzhin
3384ede4bd
simple-test: use losetup+journal_path=/dev/loopX for mounting with journal
2014-10-02 02:20:37 +04:00
Azat Khuzhin
cb833cd591
Handle bdevs for journal_path= mount flag
2014-10-02 02:20:37 +04:00
Azat Khuzhin
87e3751725
Actually load journal by calling jbd2_journal_load()
2014-10-02 02:20:37 +04:00
Azat Khuzhin
e21bfc88e3
simple-test: mount fs with journal
2014-10-02 02:20:37 +04:00
Azat Khuzhin
33f5f117df
Implement journal_path option
2014-10-02 02:20:37 +04:00
Azat Khuzhin
fd1dc7a951
Add loader for internal journal (we need to update mkfs to support this)
...
For internal journal we need enhance mkfs to use ext2fs lib, that will
allow us to create journal (with proper header).
2014-10-02 02:20:37 +04:00
Azat Khuzhin
bd1188801c
Add sfs_trace for lookup and compare inside it
2014-10-02 02:20:37 +04:00
Azat Khuzhin
65ab062528
simple-test: enable jbd2-debug
2014-10-02 02:20:37 +04:00
Azat Khuzhin
a619accdc7
Check that journal was initialized successfully
...
Otherwise you will next trace:
[ 964.775061] The magic number obtained in disk is: [268640275]
[ 964.775554] simplefs filesystem of version [1] formatted with a block size of [4096] detected in the device.
[ 964.776308] Loading journal devnum: 63
[ 964.776724] Journal device is: unknown-block(0,63)
[ 964.779494] simplefs superblock is destroyed. Unmount succesful.
[ 964.781366] Freeing private data of inode ffff88007babe3a8 (1)
[ 964.781903] simplefs is succesfully mounted on [/dev/loop0]
[ 964.782393] BUG: unable to handle kernel NULL pointer dereference at 0000000000000069
[ 964.783381] IP: [<ffffffff8112f0b3>] mount_fs+0x71/0x148
[ 964.783977] PGD 7a545067 PUD 7a514067 PMD 0
[ 964.784814] Oops: 0000 [#1 ] PREEMPT SMP
[ 964.785176] Modules linked in: simplefs(PO) loop fuse joydev hid_generic usbhid hid snd_pcm microcode snd_timer snd psmouse soundcore evdev pcspkr serio_raw ohci_pci ohci_hcd ehci_hcd processor parport_pc usbcore button battery ac thermal_sys parport i2c_piix4 usb_common i2c_core ext4 crc16 jbd2 mbcache sg sd_mod sr_mod crc_t10dif cdrom crct10dif_common ata_generic ahci libahci ata_piix libata scsi_mod e1000 [last unloaded: simplefs]
[ 964.785176] CPU: 0 PID: 3507 Comm: mount Tainted: P O 3.14.0-rc6+ #101
[ 964.785176] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
[ 964.785176] task: ffff88007cb8c6d0 ti: ffff88007ad04000 task.ti: ffff88007ad04000
[ 964.785176] RIP: 0010:[<ffffffff8112f0b3>] [<ffffffff8112f0b3>] mount_fs+0x71/0x148
[ 964.785176] RSP: 0018:ffff88007ad05e40 EFLAGS: 00010203
[ 964.785176] RAX: 0000000000000001 RBX: ffff88007c0fd000 RCX: 0000000000000007
[ 964.785176] RDX: 0000000000000001 RSI: 0000000000000046 RDI: 00000000ffffffff
[ 964.785176] RBP: ffff88007ad05e70 R08: 0000000000000004 R09: ffff8800000bbca0
[ 964.785176] R10: fffffff000007bf4 R11: 0000000000000000 R12: ffff88007a329000
[ 964.785176] R13: 0000000000000001 R14: ffffffffa01d2000 R15: 0000000000000000
[ 964.785176] FS: 00007f195521f840(0000) GS:ffff88007fc00000(0000) knlGS:0000000000000000
[ 964.785176] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 964.785176] CR2: 0000000000000069 CR3: 000000007a540000 CR4: 00000000000006f0
[ 964.785176] Stack:
[ 964.785176] ffff88007a1de080 ffff88007a1de080 ffffffffa01d2000 ffff8800377b2220
[ 964.785176] ffff88007c0fd000 ffffffffa01d2000 ffff88007ad05ea8 ffffffff81143628
[ 964.785176] 0000000081142d1b ffff88007c0fd000 ffff8800377b2220 0000000000000027
[ 964.785176] Call Trace:
[ 964.785176] [<ffffffff81143628>] vfs_kern_mount+0x5f/0xd3
[ 964.785176] [<ffffffff811458e9>] do_mount+0x6f9/0x7f0
[ 964.785176] [<ffffffff81145c01>] SyS_mount+0x80/0xb9
[ 964.785176] [<ffffffff813d2262>] system_call_fastpath+0x16/0x1b
[ 964.785176] Code: cf 06 00 85 c0 0f 85 d0 00 00 00 4c 89 ea 48 89 d9 44 89 fe 4c 89 f7 41 ff 56 10 49 89 c5 49 81 fd 00 f0 ff ff 0f 87 b0 00 00 00 <49> 8b 5d 68 48 85 db 75 02 0f 0b 48 83 bb d0 00 00 00 00 75 11
[ 964.785176] RIP [<ffffffff8112f0b3>] mount_fs+0x71/0x148
[ 964.785176] RSP <ffff88007ad05e40>
[ 964.785176] CR2: 0000000000000069
[ 964.820009] ---[ end trace fc63610f0693bff7 ]---
2014-10-02 02:20:37 +04:00
Azat Khuzhin
aa630d8e0f
Write journal device name
2014-10-02 02:20:37 +04:00
Azat Khuzhin
57a1e15c0f
Implement external journal device instead of internal
2014-10-02 02:20:37 +04:00
Azat Khuzhin
f8009d2223
simplefs_write: add journaling using jbd2
2014-10-02 02:20:37 +04:00
Azat Khuzhin
dad5b1c50e
Initial support of journaling
...
- Reorder block numbers to make it more sequental for mkfs
- mkfs: adjust writing, to make it correct write journal blocks and ino too
- simplefs_iget: use fops for journal inode
- simplefs_iget: check sfs_inode->mode instead of i_mode (not initialized yet)
2014-10-02 02:20:37 +04:00
Azat Khuzhin
072ef5bff6
Add link to build environment (github.com/azat/simplefs-bld)
2014-09-23 15:13:51 +04:00
Azat Khuzhin
8567b0e113
simple-test: call umount from cleanup() only when mount point exists
...
this will avoid errors when cleanup() called before doing actuall job,
IOW just after make and before all tests.
2014-09-17 13:38:57 +04:00
Azat Khuzhin
0a7059cd35
simple-test: use 'umount -t /path/to/mnt' to avoid umounting extra mnts
2014-09-17 13:34:51 +04:00
Azat Khuzhin
37a08dbe63
simple-test: change dir to root before unmount, to avoid opened fds
2014-09-17 13:33:37 +04:00
Azat Khuzhin
0a41a032a4
simple-test: fix awk syntax for unmount only by fstype
...
But we could just use 'umount -t' I guess.
2014-09-17 13:32:39 +04:00
Azat Khuzhin
b84270ee83
simple-test: grep out by 3 column (fs type) from /proc/mounts
2014-09-16 20:18:59 +04:00
Azat Khuzhin
e1df255a33
simple-test: more accurate grep for /proc/mounts
2014-09-16 16:22:00 +04:00
Azat Khuzhin
0fb2c3b7c1
mkfs-simplefs: fill sb by zeros sb padding
...
This will make debugging more cleaner (you will see actual zeros instead
of trash)
2014-07-12 22:26:28 +04:00
Azat Khuzhin
7610120b10
mkfs: split into more helpers (and drop malloc just do seek instead)
2014-07-12 22:26:14 +04:00
Azat Khuzhin
bb70d76497
Add BUG_ON() after each sb_bread() to avoid NULL dereferencing
...
If mkfs-simple will have bugs, than it could obtain illegal block number, and
sb_bread() will return NULL, even without it could return NULL, so since this
fs is just for testing, simply add bug_on() to avoid going further.
[ 1247.861163] The magic number obtained in disk is: [268640275]
[ 1247.861165] simplefs filesystem of version [1] formatted with a block size of [4096] detected in the device.
[ 1247.861194] simplefs is succesfully mounted on [/dev/loop0]
[ 1247.886800] BUG: unable to handle kernel NULL pointer dereference at (null)
[ 1247.886804] IP: [<ffffffffa00ca695>] simplefs_lookup+0xb5/0x180 [simplefs]
[ 1247.886808] PGD 5eda6067 PUD 5eda5067 PMD 0
[ 1247.886810] Oops: 0000 [#1 ] SMP
[ 1247.886812] Modules linked in: simplefs(POE) snd_intel8x0 snd_ac97_codec ac97_bus snd_pcm snd_seq_midi snd_seq_midi_event snd_rawmidi snd_seq snd_seq_device intel_rapl snd_timer joydev snd rfcomm bnep bluetooth 6lowpan_iphc serio_raw i2c_piix4 mac_hid soundcore parport_pc ppdev lp parport hid_generic usbhid hid psmouse ahci libahci e1000 pata_acpi [last unloaded: simplefs]
[ 1247.886826] CPU: 0 PID: 9467 Comm: cat Tainted: P OE 3.15.0-6-generic #11-Ubuntu
[ 1247.886827] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
[ 1247.886828] task: ffff88005e72cb90 ti: ffff88005ede8000 task.ti: ffff88005ede8000
[ 1247.886829] RIP: 0010:[<ffffffffa00ca695>] [<ffffffffa00ca695>] simplefs_lookup+0xb5/0x180 [simplefs]
[ 1247.886832] RSP: 0018:ffff88005edebc88 EFLAGS: 00010292
[ 1247.886833] RAX: 0000000000000002 RBX: 0000000000000000 RCX: ffff88005f5370a0
[ 1247.886834] RDX: ffff880024c9a108 RSI: ffff880024c9bb60 RDI: ffff880024c9a000
[ 1247.886834] RBP: ffff88005edebcc0 R08: 0000000000000000 R09: 0000000000000001
[ 1247.886835] R10: ffff88005edebab8 R11: 0000000000000000 R12: ffff88005f537000
[ 1247.886836] R13: ffff88005ed68000 R14: 0000000000000000 R15: ffff880024c9a000
[ 1247.886837] FS: 00007ffb35497740(0000) GS:ffff880065c00000(0000) knlGS:0000000000000000
[ 1247.886838] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1247.886839] CR2: 0000000000000000 CR3: 000000005edfa000 CR4: 00000000000006f0
[ 1247.886843] Stack:
[ 1247.886844] ffff880024c9bb60 ffff88001d19e180 ffff88001d19e180 0000000000008000
[ 1247.886845] ffff88005edebde8 ffff88001d159e40 ffff88005edebe40 ffff88005edebce0
[ 1247.886847] ffffffff811da5ed ffff88001d159e40 ffff88005edebf2c ffff88005edebd88
[ 1247.886848] Call Trace:
[ 1247.886853] [<ffffffff811da5ed>] lookup_real+0x1d/0x70
[ 1247.886856] [<ffffffff811df537>] do_last+0x797/0x1270
[ 1247.886860] [<ffffffff8132c81b>] ? apparmor_file_alloc_security+0x5b/0x180
[ 1247.886862] [<ffffffff812ed116>] ? security_file_alloc+0x16/0x20
[ 1247.886864] [<ffffffff811e119b>] path_openat+0xbb/0x6a0
[ 1247.886866] [<ffffffff811e285a>] do_filp_open+0x3a/0xa0
[ 1247.886868] [<ffffffff811ef597>] ? __alloc_fd+0xa7/0x130
[ 1247.886871] [<ffffffff811cff8a>] do_sys_open+0x12a/0x280
[ 1247.886874] [<ffffffff81020925>] ? syscall_trace_enter+0x165/0x2a0
[ 1247.886876] [<ffffffff811d00fe>] SyS_open+0x1e/0x20
[ 1247.886879] [<ffffffff8176e33f>] tracesys+0xe1/0xe6
[ 1247.886880] Code: 01 00 00 4c 89 e7 e8 ab fe ff ff 4c 89 e7 48 89 c3 e8 f0 38 12 e1 49 89 c7 49 8b 85 00 01 00 00 48 8b 34 24 4c 89 ff 49 89 47 40 <0f> b7 13 e8 83 0f 12 e1 41 0f b7 07 4d 89 67 28 49 c7 47 20 40
[ 1247.886894] RIP [<ffffffffa00ca695>] simplefs_lookup+0xb5/0x180 [simplefs]
[ 1247.886896] RSP <ffff88005edebc88>
[ 1247.886896] CR2: 0000000000000000
[ 1247.886898] ---[ end trace 89702ff0fda32679 ]---
2014-07-12 22:21:49 +04:00
Sankar சங்கர்
069f34ebbc
Merge pull request #3 from asfdfdfd/patch-1
...
Typo in comment fixed.
2014-03-29 21:15:55 +05:30
asfdfdfd
ed7ff8a42d
Typo in comment fixed.
2014-03-29 16:08:08 +03:00
Azat Khuzhin
3050cdf2ed
Use slab allocator for simplefs_inode
...
Struct simplefs_inode allocation is quite frequently (for fs), so it
will be better to use slab allocator instead of page allocator.
2013-09-30 22:46:56 +04:00
Azat Khuzhin
031f974c36
fix another inode leaking in simplefs_create_fs_object()
...
There we iterate over inodes using simplefs_get_inode(), while instead
we already have last inode number, just use it instead, and increment
some paddings to it (start inode, reserved inodes).
And form after this patch kedr shows that there is no leaks.
(but we must recheck it in the future.)
2013-09-30 02:34:05 +04:00
Azat Khuzhin
98bbe99632
Fix leaking of inode memory.
...
Release bh in simplefs_get_inode() and use kmalloc() instead, and also
add super_operations.destory_inode handler, to free allocated memory.
This is not optimal and it must be replaced by slab allocators.
2013-09-30 02:13:24 +04:00
Azat Khuzhin
7130573869
read: drop static "done" variable, we already have file_size in inode.
2013-09-29 20:22:03 +04:00
Azat Khuzhin
d976c4f65e
test: read file after truncate it to more smaller size.
...
Just write less bytes.
2013-09-29 20:18:49 +04:00
Azat Khuzhin
d67a2974e5
test: add some read operations after unmount/mount
2013-09-29 20:12:51 +04:00
Azat Khuzhin
86d7a8ff80
Add new helper - simplefs_inode_save()
...
And use it next functions:
- simplefs_write()
- simplefs_create_fs_object()
2013-09-29 20:01:35 +04:00
Azat Khuzhin
2d95ccf1a3
Start implementing verbose logging depends on macros. (SIMPLEFS_DEBUG)
...
In many cases we don't need such verbose loggin, just add macros to
print loggin only if SIMPLEFS_DEBUG isset.
And it will be installed by default. (in Makefile)
2013-09-29 18:09:44 +04:00
Azat Khuzhin
f9aa71628d
Add new helper simplefs_inode_search()
2013-09-29 16:41:48 +04:00
Azat Khuzhin
c199cecbce
simplefs_create_fs_object(): release mutexes when we can't acquire sb_lock
...
We need more pretty code for this.
2013-09-25 18:09:01 +04:00
Azat Khuzhin
ee1efa11df
sb: use s_maxbytes field instead of hardcoded check in simplefs_write()
...
generic_write_checks() will do this for us now, using s_maxbytes field
from sb.
2013-09-25 15:41:39 +04:00
Azat Khuzhin
ab68a5176a
simplefs_write(): use generic_write_checks()
2013-09-25 15:36:01 +04:00
Azat Khuzhin
274a98ae14
Add TODO about moving version depend code into separate header.
2013-09-25 00:48:31 +04:00
Azat Khuzhin
f72fbd4db6
Add new const SIMPLEFS_START_INO instead of magic number
2013-09-18 23:43:50 +04:00
Azat Khuzhin
f61cf26333
simple.h: drop trailing whitespaces
2013-09-18 23:43:17 +04:00
Azat Khuzhin
82a2041957
simplefs_create_fs_object(): avoid extra allocation of memory
...
We can just use already existed bh->b_data for this, and copy filename
and inode number directly to it.
2013-09-18 23:17:42 +04:00
Azat Khuzhin
0e14ca0edc
simplefs_fill_super() drop deprecated FIXME (see 9aee1ce)
2013-09-18 23:07:11 +04:00