On Thu, May 05, 2016 at 05:53:14PM +, Stephen Bates wrote:
> > >
> > > One thing I did notice is that fallocate() seems slow (5-6 GB/s) compared
> > > to
> > other file systems for a 3TiB fallocate() [ext4 performs the same operation
> > in
> > under a second on my system)]. Is this typical/e
This patch enables reading node blocks in advance when truncating large
data blocks.
> time rm $MNT/testfile (500GB) after drop_cachees
Before : 9.422 s
After : 4.821 s
Reported-by: Stephen Bates
Signed-off-by: Jaegeuk Kim
---
fs/f2fs/file.c | 2 +-
fs/f2fs/node.c | 52 +
Previously f2fs_preallocate_blocks() tries to allocate unaligned blocks.
In f2fs_write_begin(), however, prepare_write_begin() does not skip its
allocation due to (len != 4KB).
So, it needs locking node page twice unexpectedly.
Signed-off-by: Jaegeuk Kim
---
fs/f2fs/data.c | 4 ++--
1 file chang
Hi Chao,
Good catch, but I've been already testing a patch which uses f2fs_map_blocks().
We don't need to add whole things redundantly. :)
>From b150a6e02785ea28a5139bd2d1a1debd8aad84e7 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim
Date: Fri, 6 May 2016 15:30:38 -0700
Subject: [PATCH] f2fs: falloca
Mike Christie writes:
> On 05/03/2016 03:44 PM, Jeff Moyer wrote:
>> Hi, Mike,
>>
>> That git tree doesn't seem to exist. I did manage to apply your patch
>> set on top of next-20160415, though.
>>
>> So... what testing did you do? ;-) I ran into the following problems
>
> I normally run xfste
On Sun, May 01, 2016 at 08:19:44AM +1000, NeilBrown wrote:
> On Sat, Apr 30 2016, Dave Chinner wrote:
> > Indeed, blocking the superblock shrinker in reclaim is a key part of
> > balancing inode cache pressure in XFS. If the shrinker starts
> > hitting dirty inodes, it blocks on cleaning them, ther
On 05/03/2016 03:44 PM, Jeff Moyer wrote:
> mchri...@redhat.com writes:
>
>> The following patches begin to cleanup the request->cmd_flags and
>> bio->bi_rw mess. We currently use cmd_flags to specify the operation,
>> attributes and state of the request. For bi_rw we use it for similar
>> info an
On 05/04/2016 12:58 PM, Jeff Moyer wrote:
> Mike Christie writes:
>
>> On 05/03/2016 03:44 PM, Jeff Moyer wrote:
>>> Hi, Mike,
>>>
>>> That git tree doesn't seem to exist. I did manage to apply your patch
>>> set on top of next-20160415, though.
>>>
>>> So... what testing did you do? ;-) I ran i
Preallocation operation in ->fallocate seems quite slow, since for all
new preallocated blocks, f2fs will update them one by one in direct nodes,
This patch introduces f2fs_reserve_blocks to make all preallocated blocks
belongs to one direct node updating in batch, so it can save a lot of
cpu cycl
When testing f2fs with inline_dentry option, generic/342 reports:
VFS: Busy inodes after unmount of dm-0. Self-destruct in 5 seconds. Have a
nice day...
After rmmod f2fs module, kenrel shows following dmesg:
=
BUG f2fs
10 matches
Mail list logo
