Currently if there exists a route that goes over a
remote net and then this net is added dynamically as
a local net, then traffic stops because the code in
lnet_send() determines that the destination nid
can be reached from another local_ni, but the src_nid
is still stuck on the earlier NI, because the src_nid
is stored in the ptlrpc layer and is not updated
when a local NI is configured.
Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5874
Reviewed-on: http://review.whamcloud.com/12912
Reviewed-by: Isaac Huang <he.huang@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With DLC it's possible to start up a system with no NIs that require
the acceptor thread, and thus it won't start. Later on the user
can add an NI that requires the acceptor thread to start, it is
then necessary to start it up.
If the user removes a NI and as a result there are no more
NIs that require the acceptor thread then it should be stopped.
This patch adds logic in the dynamically adding and removing NIs
code to ensure the above logic is implemented.
Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6002
Reviewed-on: http://review.whamcloud.com/13010
Reviewed-by: Isaac Huang <he.huang@intel.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It is preferable to use NULL instead of 0 for pointers. This fixes sparse
warnings such as:
lustre/fld/fld_request.c:126:17: warning: Using plain integer as NULL pointer
The second parameter of class_match_param() was changed to a const, to
be able to remove a cast in one user, to prevent splitting a long
line. No other code change.
Signed-off-by: Frank Zago <fzago@cray.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5396
Reviewed-on: http://review.whamcloud.com/12567
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some functions and variables are only used in their C file, so reduce
their scope. This reduces the code size, and fixes sparse warnings
such as:
warning: symbol 'proc_lnet_routes' was not declared.
Should it be static?
warning: symbol 'proc_lnet_routers' was not declared.
Should it be static?
Some prototypes were removed from C files and added to the proper
header.
Signed-off-by: Frank Zago <fzago@cray.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5396
Reviewed-on: http://review.whamcloud.com/12206
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When adding route it ignored specific scenarios, namely:
1. route already exists
2. route is on a local net
3. route is unreacheable
This patch returns the appropriate return codes from the lower level
function lnet_add_route(), and then ignores the above case from the
calling function, lnet_parse_route(). This is needed so we don't
halt processing routes in the module parameters.
However, we can now add routes dynamically, and it should be returned
to the user whether adding the requested route succeeded or failed.
In userspace it is determined whether to continue adding routes or to
halt processing. Currently "lnetctl import < config" continues
adding the rest of the configuration and reports at the end which
operations passed and which ones failed.
Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6045
Reviewed-on: http://review.whamcloud.com/13116
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Isaac Huang <he.huang@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch addresses a set of related issues: LU-5568, LU-5734,
LU-5839, LU-5849, LU-5850.
Create the local lnet_startup_lndni() API. This function starts
up one LND. lnet_startup_lndnis() calls this function in a loop
on every ni in the list passed in. lnet_startup_lndni() is
responsible for cleaning up after itself in case of failure.
It calls lnet_free_ni() if the ni fails to start. It calls
lnet_shutdown_lndni() if it successfully called the
lnd startup function, but fails later on.
lnet_startup_lndnis() also cleans up after itself.
If lnet_startup_lndni() fails then lnet_shutdown_lndnis() is
called to clean up all nis that might have been
started, and then free the rest of the nis on the list
which have not been started yet.
To facilitate the above changes lnet_dyn_del_ni() now
manages the ping info. It calls lnet_shutdown_lndni(),
to shutdown the NI. lnet_shutdown_lndni() is no longer
an exposed API and doesn't manage the ping info, making
it callable from lnet_startup_lndni() as well.
There are two scenarios for calling lnet_startup_lndni()
1. from lnet_startup_lndnis()
If lnet_startup_lndni() fails it requires to shutdown the ni
without doing anything with the ping information as it hasn't
been created yet.
2. from lnet_dyn_add_ni()
As above it will shutdown the ni, and then lnet_dyn_add_ni() will
take care of managing the ping info
The second part of this change is to ensure that the LOLND is not
added by lnet_parse_networks(), but the caller which needs to do
it (IE: LNetNIInit)
This change ensures that lnet_dyn_add_ni() need only check if there is
only one net that's being added, if not then it frees everything,
otherwise it proceeds to startup the requested net.
Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5734
Reviewed-on: http://review.whamcloud.com/12658
Reviewed-by: Liang Zhen <liang.zhen@intel.com>
Reviewed-by: Isaac Huang <he.huang@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove LUSTRE_LNET_PID (12354) and LUSTRE_SRV_LNET_PID (12345) from
the libcfs headers and replace their uses with a new macro
LNET_PID_LUSTRE (also 12345) in lnet/types.h.
Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2675
Reviewed-on: http://review.whamcloud.com/11985
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The function lnet_unprepare can be made static.
Signed-off-by: Frank Zago <fzago@cray.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5396
Reviewed-on: http://review.whamcloud.com/11306
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Patrick Farrell <paf@cray.com>
Reviewed-by: Amir Shehata <amir.shehata@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is the sixth patch of a set of patches that enables DLC.
This patch enables the user space to call into the kernel space
DLC code. Added handlers in the LNetCtl function to call
the new functions added for Dynamic Lnet Configuration
Signed-off-by: Amir Shehata <amir.shehata@intel.com>
ntel-bug-id: https://jira.hpdd.intel.com/browse/LU-2456
Reviewed-on: http://review.whamcloud.com/8023
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If there is an invalid networks or ip2nets lnet_parse_networks()
gets called with a NULL 'network' string parameter
lnet_parse_networks() needs to sanitize its input string now that
it's being called from multiple places. Instead, check for
a NULL string everytime the function is called, which reduces the
probability of errors with other code modifications.
Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5540
Reviewed-on: http://review.whamcloud.com/11626
Reviewed-by: Isaac Huang <he.huang@intel.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is the fifth patch of a set of patches that enables DLC.
This patch adds the new structures which will be used
in the IOCTL communication. It also added a set of
show operations to show buffers, networks, statistics
and peer information.
Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2456
Reviewed-on: http://review.whamcloud.com/8022
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is the fourth patch of a set of patches that enables DLC.
This patch changes the IOCTL infrastructure in preparation of
adding extra IOCTL communication between user and kernel space.
The changes include:
- adding a common header to be passed to ioctl infra functions
instead of passing an exact structure. This header is meant
to be included in all structures to be passed through that
interface. The IOCTL handler casts this header to a particular
type that it expects
- All sanity testing on the past in structure is performed in the
generic ioctl infrastructure code.
- All ioctl handlers changed to take the header instead of a
particular structure type
Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2456
Reviewed-on: http://review.whamcloud.com/8021
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: James Simmons <uja.ornl@gmail.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove unimplemented svc unique_id attribute from the documentation.
This attribute made more sense when we thought we'd have an AP-module,
unlike now when the AP and SVC are both part of the same frame.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
It might be of interest (to developers at least) to know when an
interface is getting created or removed from the system.
Interface creation message can further contain basic information about
the interface, like its vid/pid and mfg/prod ids.
Now, the interface is created by gb_interface_create(), which doesn't
register the intf->dev to the kernel and so the print message is rather
added to gb_interface_init() where we register the device with the
kernel.
A similar message is added to gb_interface_remove() only when the
interface was earlier initialized.
And this is how the output looks on real insertion/removal of the
module:
greybus 1-1: Interface added: VID=0x00000001, PID=0x00000001
greybus 1-1: DDBL1 Manufacturer=0x00000001, Product=0x00000001
...
greybus 1-1: Interface removed
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Reviewed-by: Johan Hovold <johan@hovoldconsulting.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
The intel-pstate driver is using intel_pstate_hwp_set() from two
separate paths, i.e. ->set_policy() callback and sysfs update path for
the files present in /sys/devices/system/cpu/intel_pstate/ directory.
While an update to the sysfs path applies to all the CPUs being managed
by the driver (which essentially means all the online CPUs), the update
via the ->set_policy() callback applies to a smaller group of CPUs
managed by the policy for which ->set_policy() is called.
And so, intel_pstate_hwp_set() should update frequencies of only the
CPUs that are part of policy->cpus mask, while it is called from
->set_policy() callback.
In order to do that, add a parameter (cpumask) to intel_pstate_hwp_set()
and apply the frequency changes only to the concerned CPUs.
For ->set_policy() path, we are only concerned about policy->cpus, and
so policy->rwsem lock taken by the core prior to calling ->set_policy()
is enough to take care of any races. The larger lock acquired by
get_online_cpus() is required only for the updates to sysfs files.
Add another routine, intel_pstate_hwp_set_online_cpus(), and call it
from the sysfs update paths.
This also fixes a lockdep reported recently, where policy->rwsem and
get_online_cpus() could have been acquired in any order causing an ABBA
deadlock. The sequence of events leading to that was:
intel_pstate_init(...)
...cpufreq_online(...)
down_write(&policy->rwsem); // Locks policy->rwsem
...
cpufreq_init_policy(policy);
...intel_pstate_hwp_set();
get_online_cpus(); // Temporarily locks cpu_hotplug.lock
...
up_write(&policy->rwsem);
pm_suspend(...)
...disable_nonboot_cpus()
_cpu_down()
cpu_hotplug_begin(); // Locks cpu_hotplug.lock
__cpu_notify(CPU_DOWN_PREPARE, ...);
...cpufreq_offline_prepare();
down_write(&policy->rwsem); // Locks policy->rwsem
Reported-and-tested-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Acked-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
This patch adopts f2fs with codes of ext4, it removes unneeded memory
allocation in creating/accessing path of symlink.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch syncs f2fs with commit abdd438b26 ("ext4 crypto: handle
unexpected lack of encryption keys") from ext4.
Fix up attempts by users to try to write to a file when they don't
have access to the encryption key.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch syncs f2fs with commit 6bc445e0ff ("ext4 crypto: make
sure the encryption info is initialized on opendir(2)") from ext4.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
f2fs support atomic write with following semantics:
1. open db file
2. ioctl start atomic write
3. (write db file) * n
4. ioctl commit atomic write
5. close db file
With this flow we can avoid file becoming corrupted when abnormal power
cut, because we hold data of transaction in referenced pages linked in
inmem_pages list of inode, but without setting them dirty, so these data
won't be persisted unless we commit them in step 4.
But we should still hold journal db file in memory by using volatile
write, because our semantics of 'atomic write support' is incomplete, in
step 4, we could fail to submit all dirty data of transaction, once
partial dirty data was committed in storage, then after a checkpoint &
abnormal power-cut, db file will be corrupted forever.
So this patch tries to improve atomic write flow by adding a revoking flow,
once inner error occurs in committing, this gives another chance to try to
revoke these partial submitted data of current transaction, it makes
committing operation more like aotmical one.
If we're not lucky, once revoking operation was failed, EAGAIN will be
reported to user for suggesting doing the recovery with held journal file,
or retrying current transaction again.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Split drop_inmem_pages from commit_inmem_pages for code readability,
and prepare for the following modification.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch fixes to eliminate garbage name lengths in dentries in order
to provide correct answers of readdir.
For example, if a valid dentry consists of:
bitmap : 1 1 1 1
len : 32 0 x 0,
readdir can start with second bit_pos having len = 0.
Or, it can start with third bit_pos having garbage.
In both of cases, we should avoid to try filling dentries.
So, this patch not only removes any garbage length, but also avoid entering
zero length case in readdir.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch adopts:
ext4 crypto: check for too-short encrypted file names
An encrypted file name should never be shorter than an 16 bytes, the
AES block size. The 3.10 crypto layer will oops and crash the kernel
if ciphertext shorter than the block size is passed to it.
Fortunately, in modern kernels the crypto layer will not crash the
kernel in this scenario, but nevertheless, it represents a corrupted
directory, and we should detect it and mark the file system as
corrupted so that e2fsck can fix this.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch adopts:
ext4 crypto: ext4_page_crypto() doesn't need a encryption context
Since ext4_page_crypto() doesn't need an encryption context (at least
not any more), this allows us to simplify a number function signature
and also allows us to avoid needing to allocate a context in
ext4_block_write_begin(). It also means we no longer need a separate
ext4_decrypt_one() function.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch adopts:
ext4 crypto: replace some BUG_ON()'s with error checks
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch preallocates data blocks for buffered aio writes.
With this patch, we can avoid redundant locking and unlocking of node pages
given consecutive aio request.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
fix missing skip pages info in f2fs_writepages trace event.
Signed-off-by: Yunlei He <heyunlei@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
f2fs use single bio buffer per type data (META/NODE/DATA) for caching
writes locating in continuous block address as many as possible, after
submitting, these writes may be still cached in bio buffer, so we have
to flush cached writes in bio buffer by calling f2fs_submit_merged_bio.
Unfortunately, in the scenario of high concurrency, bio buffer could be
flushed by someone else before we submit it as below reasons:
a) there is no space in bio buffer.
b) add a request of different type (SYNC, ASYNC).
c) add a discontinuous block address.
For this condition, f2fs_submit_merged_bio will be devastating, because
it could break the following merging of writes in bio buffer, split one
big bio into two smaller one.
This patch introduces f2fs_submit_merged_bio_cond which can do a
conditional submitting with bio buffer, before submitting it will judge
whether:
- page in DATA type bio buffer is matching with specified page;
- page in DATA type bio buffer is belong to specified inode;
- page in NODE type bio buffer is belong to specified inode;
If there is no eligible page in bio buffer, we will skip submitting step,
result in gaining more chance to merge consecutive block IOs in bio cache.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Sometimes, if cp_error is set, there remains under-writeback pages, resulting in
kernel hang in put_super.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Likewise f2fs_write_cache_pages, let's do for node and meta pages too.
Especially, for node blocks, we should do this before marking its fsync
and dentry flags.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch makes f2fs_map_blocks supporting returning next potential
page offset which skips hole region in indirect tree of inode, and
use it to speed up fiemap in handling big hole case.
Test method:
xfs_io -f /mnt/f2fs/file -c "pwrite 1099511627776 4096"
time xfs_io -f /mnt/f2fs/file -c "fiemap -v"
Before:
time xfs_io -f /mnt/f2fs/file -c "fiemap -v"
/mnt/f2fs/file:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [0..2147483647]: hole 2147483648
1: [2147483648..2147483655]: 81920..81927 8 0x1
real 3m3.518s
user 0m0.000s
sys 3m3.456s
After:
time xfs_io -f /mnt/f2fs/file -c "fiemap -v"
/mnt/f2fs/file:
EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS
0: [0..2147483647]: hole 2147483648
1: [2147483648..2147483655]: 81920..81927 8 0x1
real 0m0.008s
user 0m0.000s
sys 0m0.008s
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
When seeking data in ->llseek, if we encounter a big hole which covers
several dnode pages, we will try to seek data from index of page which
is the first page of next dnode page, at most we could skip searching
(ADDRS_PER_BLOCK - 1) pages.
However it's still not efficient, because if our indirect/double-indirect
pointer are NULL, there are no dnode page locate in the tree indirect/
double-indirect pointer point to, it's not necessary to search the whole
region.
This patch introduces get_next_page_offset to calculate next page offset
based on current searching level and max searching level returned from
get_dnode_of_data, with this, we could skip searching the entire area
indirect or double-indirect node block is not exist.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
There are redundant pointer conversion in following call stack:
- at position a, inode was been converted to f2fs_file_info.
- at position b, f2fs_file_info was been converted to inode again.
- truncate_blocks(inode,..)
- fi = F2FS_I(inode) ---a
- ADDRS_PER_PAGE(node_page, fi)
- addrs_per_inode(fi)
- inode = &fi->vfs_inode ---b
- f2fs_has_inline_xattr(inode)
- fi = F2FS_I(inode)
- is_inode_flag_set(fi,..)
In order to avoid unneeded conversion, alter ADDRS_PER_PAGE and
addrs_per_inode to acept parameter with type of inode pointer.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch uses existing function f2fs_map_block to simplify implementation
of __allocate_data_blocks.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
In f2fs_map_blocks, we use duplicated codes to handle first block mapping
and the following blocks mapping, it's unnecessary. This patch simplifies
f2fs_map_blocks to avoid using copied codes.
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch introduces lifetime IO write statistics exposed to the sysfs interface.
The write IO amount is obtained from block layer, accumulated in the file system and
stored in the hot node summary of checkpoint.
Signed-off-by: Shuoran Liu <liushuoran@huawei.com>
Signed-off-by: Pengyang Hou <houpengyang@huawei.com>
[Jaegeuk Kim: add sysfs documentation]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
On the worst case, we need to scan the whole radix tree and every rb-tree to
free the victimed extent_nodes when shrinking.
Pengyang initially introduced a victim_list to record the victimed extent_nodes,
and free these extent_nodes by just scanning a list.
Later, Chao Yu enhances the original patch to improve memory footprint by
removing victim list.
The policy of lru list shrinking becomes:
1) lock lru list's lock
2) trylock extent tree's lock
3) remove extent node from lru list
4) unlock lru list's lock
5) do shrink
6) repeat 1) to 5)
Signed-off-by: Hou Pengyang <houpengyang@huawei.com>
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
There are three steps to free an extent node:
1) list_del_init, 2)__detach_extent_node, 3) kmem_cache_free
In path f2fs_destroy_extent_tree, 1->2->3 to free a node,
But in path f2fs_update_extent_tree_range, it is 2->1->3.
This patch makes all the order to be: 1->2->3
It makes sense, since in the next patch, we import a victim list in the
path shrink_extent_tree, we could check if the extent_node is in the victim
list by checking the list_empty(). So it is necessary to put 1) first.
Signed-off-by: Hou Pengyang <houpengyang@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
variable nsearched in get_victim_by_default() indicates the number of
dirty segments we already checked. There are 2 problems about the way
it updates:
1. When p.ofs_unit is greater than 1, the victim we find consists
of multiple segments, possibly more than 1 dirty segment.
But nsearched always increases by 1.
2. If segments have been found but not been chosen, nsearched won't
increase. So even we have checked all dirty segments, nsearched
may still less than p.max_search.
All these problems could cause unnecessary search after all dirty
segments have already been checked.
Signed-off-by: Fan li <fanofcode.li@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
