Patch series "kvmalloc", v5.

There are many open coded kmalloc with vmalloc fallback instances in the
tree.  Most of them are not careful enough or simply do not care about
the underlying semantic of the kmalloc/page allocator which means that
a) some vmalloc fallbacks are basically unreachable because the kmalloc
part will keep retrying until it succeeds b) the page allocator can
invoke a really disruptive steps like the OOM killer to move forward
which doesn't sound appropriate when we consider that the vmalloc
fallback is available.

As it can be seen implementing kvmalloc requires quite an intimate
knowledge if the page allocator and the memory reclaim internals which
strongly suggests that a helper should be implemented in the memory
subsystem proper.

Most callers, I could find, have been converted to use the helper
instead.  This is patch 6.  There are some more relying on __GFP_REPEAT
in the networking stack which I have converted as well and Eric Dumazet
was not opposed [2] to convert them as well.

[1] http://lkml.kernel.org/r/20170130094940.13546-1-mhocko@kernel.org
[2] http://lkml.kernel.org/r/1485273626.16328.301.camel@edumazet-glaptop3.roam.corp.google.com

This patch (of 9):

Using kmalloc with the vmalloc fallback for larger allocations is a
common pattern in the kernel code.  Yet we do not have any common helper
for that and so users have invented their own helpers.  Some of them are
really creative when doing so.  Let's just add kv[mz]alloc and make sure
it is implemented properly.  This implementation makes sure to not make
a large memory pressure for > PAGE_SZE requests (__GFP_NORETRY) and also
to not warn about allocation failures.  This also rules out the OOM
killer as the vmalloc is a more approapriate fallback than a disruptive
user visible action.

This patch also changes some existing users and removes helpers which
are specific for them.  In some cases this is not possible (e.g.
ext4_kvmalloc, libcfs_kvzalloc) because those seems to be broken and
require GFP_NO{FS,IO} context which is not vmalloc compatible in general
(note that the page table allocation is GFP_KERNEL).  Those need to be
fixed separately.

While we are at it, document that __vmalloc{_node} about unsupported gfp
mask because there seems to be a lot of confusion out there.
kvmalloc_node will warn about GFP_KERNEL incompatible (which are not
superset) flags to catch new abusers.  Existing ones would have to die
slowly.

[sfr@canb.auug.org.au: f2fs fixup]
  Link: http://lkml.kernel.org/r/20170320163735.332e64b7@canb.auug.org.au
Link: http://lkml.kernel.org/r/20170306103032.2540-2-mhocko@kernel.orgSigned-off-by: NMichal Hocko <mhocko@suse.com>
Signed-off-by: NStephen Rothwell <sfr@canb.auug.org.au>
Reviewed-by: Andreas Dilger <adilger@dilger.ca>	[ext4 part]
Acked-by: NVlastimil Babka <vbabka@suse.cz>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Both nat_bits cache and free_nid_bitmap cache provide same functionality
as a intermediate cache between free nid cache and disk, but with
different granularity of indicating free nid range, and different
persistence policy. nat_bits cache provides better persistence ability,
and free_nid_bitmap provides better granularity.

In this patch we combine advantage of both caches, so finally policy of
the intermediate cache would be:
- init: load free nid status from nat_bits into free_nid_bitmap
- lookup: scan free_nid_bitmap before load NAT blocks
- update: update free_nid_bitmap in real-time
- persistence: udpate and persist nat_bits in checkpoint

This patch also resolves performance regression reported by lkp-robot.

commit:
  4ac91242 ("f2fs: introduce free nid bitmap")
  d00030cf9cd0bb96fdccc41e33d3c91dcbb672ba ("f2fs: use __set{__clear}_bit_le")
  1382c0f3f9d3f936c8bc42ed1591cf7a593ef9f7 ("f2fs: combine nat_bits and free_nid_bitmap cache")

4ac91242 d00030cf9cd0bb96fdccc41e33 1382c0f3f9d3f936c8bc42ed15
---------------- -------------------------- --------------------------
         %stddev     %change         %stddev     %change         %stddev
             \          |                \          |                \
     77863 ±  0%      +2.1%      79485 ±  1%     +50.8%     117404 ±  0%  aim7.jobs-per-min
    231.63 ±  0%      -2.0%     227.01 ±  1%     -33.6%     153.80 ±  0%  aim7.time.elapsed_time
    231.63 ±  0%      -2.0%     227.01 ±  1%     -33.6%     153.80 ±  0%  aim7.time.elapsed_time.max
    896604 ±  0%      -0.8%     889221 ±  3%     -20.2%     715260 ±  1%  aim7.time.involuntary_context_switches
      2394 ±  1%      +4.6%       2503 ±  1%      +3.7%       2481 ±  2%  aim7.time.maximum_resident_set_size
      6240 ±  0%      -1.5%       6145 ±  1%     -14.1%       5360 ±  1%  aim7.time.system_time
   1111357 ±  3%      +1.9%    1132509 ±  2%      -6.2%    1041932 ±  2%  aim7.time.voluntary_context_switches
...
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Tested-by: NXiaolong Ye <xiaolong.ye@intel.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This patch adds to account free nids for each NAT blocks, and while
scanning all free nid bitmap, do check count and skip lookuping in
full NAT block.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This patch uses __set{__clear}_bit_le for highter speed.
Reviewed-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This is to avoid build warning reported by kbuild test robot.
Signed-off-by: NFengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This patch adds a missing condition which flushes nat journal entries
unnecessarily introduced by:

    f2fs: add bitmaps for empty or full NAT blocks
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

F2FS has define MAX_FREE_NIDS for maximum of cached free nids target.
Signed-off-by: NKinglong Mee <kinglongmee@gmail.com>
Reviewed-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

In scenario of intensively node allocation, free nids will be ran out
soon, then it needs to stop to load free nids by traversing NAT blocks,
in worse case, if NAT blocks does not be cached in memory, it generates
IOs which slows down our foreground operations.

In order to speed up node allocation, in this patch we introduce a new
free_nid_bitmap array, so there is an bitmap table for each NAT block,
Once the NAT block is loaded, related bitmap cache will be switched on,
and bitmap will be set during traversing nat entries in NAT block, later
we can query and update nid usage status in memory completely.

With such implementation, I expect performance of node allocation can be
improved in the long-term after filesystem image is mounted.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

There are four places that getting the crc value in f2fs_checkpoint,
just add a new helper cur_cp_crc for them.
Signed-off-by: NKinglong Mee <kinglongmee@gmail.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

Previously kernel message can show that in which function we do the
injection, but unfortunately, most of the caller are the same, for
tracking more information of injection path, it needs to show upper
caller's name. This patch supports that ability.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This patches adds bitmaps to represent empty or full NAT blocks containing
free nid entries.

If we can find valid crc|cp_ver in the last block of checkpoint pack, we'll
use these bitmaps when building free nids. In order to avoid checkpointing
burden, up-to-date bitmaps will be flushed only during umount time. So,
normally we can get this gain, but when power-cut happens, we rely on fsck.f2fs
which recovers this bitmap again.

After this patch, we build free nids from nid #0 at mount time to make more
full NAT blocks, but in runtime, we check empty NAT blocks to load free nids
without loading any NAT pages from disk.
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

We've not seen this buggy case for a long time, so it's time to avoid this
unnecessary get_node_info() call which reading NAT page to cache nat entry.
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

Currently, if we call fsync after updating the xattr date belongs to the
file, f2fs needs to trigger checkpoint to keep xattr data consistent. But,
this policy cause low performance as checkpoint will block most foreground
operations and cause unneeded and unrelated IOs around checkpoint.

This patch will reuse regular file recovery policy for xattr node block,
so, we change to write xattr node block tagged with fsync flag to warm
area instead of cold area, and during recovery, we search warm node chain
for fsynced xattr block, and do the recovery.

So, for below application IO pattern, performance can be improved
obviously:
- touch file
- create/update/delete xattr entry in file
- fsync file
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

If the cached bio has the last page's index, then we need to submit it.
Otherwise, we don't need to submit it and can wait for further IO merges.
Reviewed-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This patch check IO submission more precisely than previous rough check.
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

We need to flush data writes before flushing last node block writes by using
FUA with PREFLUSH. We don't need to guarantee precedent node writes since if
those are not written, we can't reach to the last node block when scanning
node block chain during roll-forward recovery.
Afterwards f2fs_wait_on_page_writeback guarantees all the IO submission to
disk, which builds a valid node block chain.
Reviewed-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This patch just moves write_node_page and introduces an inner function.
Reviewed-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This patch adds a mirror for nat version bitmap, and use it to detect
in-memory bitmap corruption which may be caused by bit-transition of
cache or memory overflow.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

If we run out of memory, in cache_nat_entry, it's better to avoid loop
for allocating memory to cache nat entry, so in low memory scenario, for
read path of node block, I expect this can avoid unneeded latency.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

Thread A		Thread B		Thread C
- f2fs_create
 - f2fs_new_inode
  - f2fs_lock_op
   - alloc_nid
    alloc last nid
  - f2fs_unlock_op
			- f2fs_create
			 - f2fs_new_inode
			  - f2fs_lock_op
			   - alloc_nid
			    as node count still not
			    be increased, we will
			    loop in alloc_nid
						- f2fs_write_node_pages
						 - f2fs_balance_fs_bg
						  - f2fs_sync_fs
						   - write_checkpoint
						    - block_operations
						     - f2fs_lock_all
 - f2fs_lock_op

While creating new inode, we do not allocate and account nid atomically,
so that when there is almost no free nids left, we may encounter deadloop
like above stack.

In order to avoid that, reuse nm_i::available_nids for accounting free nids
and make nid allocation and counting being atomical during node creation.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

Thread A			Thread B

- write_checkpoint
 - block_operations
   -blk_start_plug
    -sync_node_pages		- f2fs_do_sync_file
				 - fsync_node_pages
				  - f2fs_wait_on_page_writeback

Thread A wait for global F2FS_DIRTY_NODES decreased to zero,
it start a plug list, some requests have been added to this list.
Thread B lock one dirty node page, and wait this page write back.
But this page has been in plug list of thread A with PG_writeback flag.
Thread A keep on running and its plug list has no chance to finish,
so it seems a deadlock between cp and fsync path.

This patch add a wait on page write back before set node page dirty
to avoid this problem.
Signed-off-by: NYunlei He <heyunlei@huawei.com>
Signed-off-by: NPengyang Hou <houpengyang@huawei.com>
Reviewed-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

We don't need to allocate bio partially in order to maximize sequential writes.
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This patch avoids build warning.
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

Let build_free_nids support sync/async methods, in allocation flow of nids,
we use synchronuous method, so that we can avoid looping in alloc_nid when
free memory is low; in unblock_operations and f2fs_balance_fs_bg we use
asynchronuous method in where low memory condition can interrupt us.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This patch cleans up to use consistent free nid list ops.
Reviewed-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

During free nid allocation, in order to do preallocation, we will tag free
nid entry as allocated one and still leave it in free nid list, for other
allocators who want to grab free nids, it needs to traverse the free nid
list for lookup. It becomes overhead in scenario of allocating free nid
intensively by multithreads.

This patch splits free nid list to two list: {free,alloc}_nid_list, to
keep free nids and preallocated free nids separately, after that, traverse
latency will be gone, besides split nid_cnt for separate statistic.

Additionally, introduce __insert_nid_to_list and __remove_nid_from_list for
cleanup.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: modify f2fs_bug_on to avoid needless branches]
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

f2fs contained a number of endianness conversion bugs.

Also, one function should have been 'static'.

Found with sparse by running 'make C=2 CF=-D__CHECK_ENDIAN__ fs/f2fs/'
Signed-off-by: NEric Biggers <ebiggers@google.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

In fsync_node_pages, if f2fs was taged with CP_ERROR_FLAG, make sure bio
cache was flushed before return.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

If there is no dirty pages in inode, we should give a chance to detach
the inode from global dirty list, otherwise it needs to call another
unnecessary .writepages for detaching.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

During nid allocation, it needs to exclude building and allocating flow
of free nids, this is because while building free nid cache, there are two
steps: a) load free nids from unused nat entries in NAT pages, b) update
free nid cache by checking nat journal. The two steps should be atomical,
otherwise an used nid can be allocated as free one after a) and before b).

This patch adds missing lock which covers build_free_nids in
unlock_operation and f2fs_balance_fs_bg to avoid that.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

Add wbc_to_write_flags(), which returns the write modifier flags to use,
based on a struct writeback_control. No functional changes in this
patch, but it prepares us for factoring other wbc fields for write type.
Signed-off-by: NJens Axboe <axboe@fb.com>
Reviewed-by: NJan Kara <jack@suse.cz>
Reviewed-by: NChristoph Hellwig <hch@lst.de>

Remove the WRITE_* and READ_SYNC wrappers, and just use the flags
directly.  Where applicable this also drops usage of the
bio_set_op_attrs wrapper.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <axboe@fb.com>

In sync_node_pages, we won't check and commit last merged pages in private
bio cache of f2fs, as these pages were taged as writeback, someone who is
waiting for writebacking of the page will be blocked until the cache was
committed by someone else.

We need to commit node type bio cache to avoid potential deadlock or long
delay of waiting writeback.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

Previously, we only support global fault injection configuration, so that
when we configure type/rate of fault injection through sysfs, mount
option, it will influence all f2fs partition which is being used.

It is not make sence, since it will be not convenient if developer want
to test separated partitions with different fault injection rate/type
simultaneously, also it's not possible to enable fault injection in one
partition and disable fault injection in other one.

>From now on, we move global configuration of fault injection in module
into per-superblock, hence injection testing can be more flexible.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

This patch improves the migration of dirty pages and allows migrating atomic
written pages that F2FS uses in Page Cache. Instead of the fallback releasing
page path, it provides better performance for memory compaction, CMA and other
users of memory page migrating. For dirty pages, there is no need to write back
first when migrating. For an atomic written page before committing, we can
migrate the page and update the related 'inmem_pages' list at the same time.
Signed-off-by: NWeichao Guo <guoweichao@huawei.com>
Reviewed-by: NChao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: fix some coding style]
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

Signed-off-by: NMiklos Szeredi <mszeredi@redhat.com>
Cc: Jaegeuk Kim <jaegeuk@kernel.org>

This patch gives another chances during roll-forward recovery regarding to
-ENOMEM.
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

In available_free_memory, there are two same judgement conditions which
is used for checking NAT excess, remove one of them.
Signed-off-by: NChao Yu <yuchao0@huawei.com>
Signed-off-by: NJaegeuk Kim <jaegeuk@kernel.org>

LKP reported -36.3% regression of fsmark.files_per_sec due to this patch.
I've confirmed that fxmark [1] has also slight regression for DWAL.

[1] https://github.com/sslab-gatech/fxmark

This reverts commit ec795418.

These two are confusing leftover of the old world order, combining
values of the REQ_OP_ and REQ_ namespaces.  For callers that don't
special case we mostly just replace bi_rw with bio_data_dir or
op_is_write, except for the few cases where a switch over the REQ_OP_
values makes more sense.  Any check for READA is replaced with an
explicit check for REQ_RAHEAD.  Also remove the READA alias for
REQ_RAHEAD.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJohannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: NMike Christie <mchristi@redhat.com>
Signed-off-by: NJens Axboe <axboe@fb.com>