* Kernel panic in F2FS mount on NVMe SSD.
@ 2016-05-04 22:47 Stephen Bates
2016-05-05 0:09 ` Jaegeuk Kim
0 siblings, 1 reply; 9+ messages in thread
From: Stephen Bates @ 2016-05-04 22:47 UTC (permalink / raw)
To: linux-f2fs-devel
Hi
I have been experimenting with putting F2FS onto a NVMe block device. I have been hitting a kernel panic on mount and was wondering if anyone on the list had any ideas what might be causing this. I formatted the NVMe block device using the f2fs-tools version 1.2.0 with default format settings. This is Ubuntu 14.04 running v2.6-rc6 (though I see a similar problem on older kernels).
The NVMe device is about 4TB big and is formatted for 4KB sectors. I have done some long term stability testing on the same drive using ext4 so the issue does seem to be F2FS related.
[ 328.464442]
[ 328.464449] **********************************************************
[ 328.464451] ** NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE **
[ 328.464453] ** **
[ 328.464455] ** trace_printk() being used. Allocating extra memory. **
[ 328.464456] ** **
[ 328.464458] ** This means that this is a DEBUG kernel and it is **
[ 328.464460] ** unsafe for production use. **
[ 328.464461] ** **
[ 328.464463] ** If you see this message and you are not debugging **
[ 328.464465] ** the kernel, report this immediately to your vendor! **
[ 328.464466] ** **
[ 328.464468] ** NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE **
[ 328.464470] **********************************************************
[ 330.031867] ------------[ cut here ]------------
[ 330.031962] kernel BUG at fs/f2fs/segment.h:608!
[ 330.032004] invalid opcode: 0000 [#1] SMP
[ 330.032004] Modules linked in: crc32_generic(E) f2fs(E) xt_tcpudp(E) ip6table_filter(E) ip6_tables(E) iptable_filter(E) ip_tables(E) x_tables(E) des_generic(E) md4(E) rfcomm(E) bnep(E) bluetooth(E) nls_utf8(E) cifs(E) fscache(E) amdkfd(E) amd_iommu_v2(E) radeon(E) i2c_algo_bit(E) drm_kms_helper(E) syscopyarea(E) sysfillrect(E) joydev(E) sysimgblt(E) fb_sys_fops(E) ttm(E) drm(E) nvme(E) gpio_ich(E) nvme_core(E) serio_raw(E) coretemp(E) i5000_edac(E) 8250_fintek(E) edac_core(E) lpc_ich(E) shpchp(E) mac_hid(E) i5k_amb(E) parport_pc(E) ppdev(E) lp(E) parport(E) hid_generic(E) usbhid(E) hid(E) pata_acpi(E) e1000e(E) ptp(E) pps_core(E) floppy(E)
[ 330.032004] CPU: 1 PID: 2913 Comm: mount Tainted: G E 4.6.0-rc6+nevado-v4.6-rc6 #2
[ 330.032004] Hardware name: Supermicro X7DB8/X7DB8, BIOS 2.1 06/23/2008
[ 330.032004] task: ffff8800cabe3400 ti: ffff88009f448000 task.ti: ffff88009f448000
[ 330.032004] RIP: 0010:[<ffffffffc0509e1f>] [<ffffffffc0509e1f>] build_segment_manager+0x191f/0x1980 [f2fs]
[ 330.032004] RSP: 0018:ffff88009f44bc40 EFLAGS: 00010297
[ 330.032004] RAX: 0000000000000000 RBX: 00000000001a30a1 RCX: 00000000000001fe
[ 330.032004] RDX: 0000000000000000 RSI: 0000000000000200 RDI: ffff88009f44bc80
[ 330.032004] RBP: ffff88009f44bcf8 R08: 3ffffffffe9da846 R09: ffffffffffffffff
[ 330.032004] R10: ffff88010ceeecb8 R11: ffff88009f44bb60 R12: ffffc90008177338
[ 330.032004] R13: 0000000000000072 R14: 0000000000000200 R15: ffff8800c15b1800
[ 330.032004] FS: 00007f3728b35880(0000) GS:ffff88012fc80000(0000) knlGS:0000000000000000
[ 330.032004] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 330.032004] CR2: 00007f9530eecf70 CR3: 00000000a2e23000 CR4: 00000000000006e0
[ 330.032004] Stack:
[ 330.032004] 0000000000008137 00007c0000000400 ffff8800001aa400 ffff8801292b0f00
[ 330.032004] ffff8800001a30a1 ffff8800cabb0120 ffff8800b98dd400 8000000000000000
[ 330.032004] 4000000000f7874d 3000000000432464 d000000002c71521 9000000000550b07
[ 330.032004] Call Trace:
[ 330.032004] [<ffffffffc04f2bbd>] f2fs_fill_super+0x62d/0xfe0 [f2fs]
[ 330.032004] [<ffffffff81209217>] mount_bdev+0x187/0x1c0
[ 330.032004] [<ffffffffc04f2590>] ? f2fs_commit_super+0xc0/0xc0 [f2fs]
[ 330.032004] [<ffffffffc04ecbe5>] f2fs_mount+0x15/0x20 [f2fs]
[ 330.032004] [<ffffffff81209af9>] mount_fs+0x39/0x160
[ 330.032004] [<ffffffff811a6145>] ? __alloc_percpu+0x15/0x20
[ 330.032004] [<ffffffff81224c47>] vfs_kern_mount+0x67/0x110
[ 330.032004] [<ffffffff812271f5>] do_mount+0x225/0xdb0
[ 330.032004] [<ffffffff811e542f>] ? __kmalloc_track_caller+0x1af/0x250
[ 330.032004] [<ffffffff811a0a31>] ? strndup_user+0x41/0x80
[ 330.032004] [<ffffffff811a0952>] ? memdup_user+0x42/0x70
[ 330.032004] [<ffffffff81228063>] SyS_mount+0x83/0xd0
[ 330.032004] [<ffffffff817aecf6>] entry_SYSCALL_64_fastpath+0x1e/0xa8
[ 330.032004] Code: 0e 48 83 c7 04 48 83 c6 04 83 ea 04 89 4f fc e9 8d ed ff ff 48 83 ea 01 e9 7a ef ff ff 48 8d 7a ff e9 d6 ee ff ff e8 3d 61 00 00 <0f> 0b 4c 8d 68 ff e9 af f7 ff ff 48 83 e8 01 e9 87 f7 ff ff e8
[ 330.032004] RIP [<ffffffffc0509e1f>] build_segment_manager+0x191f/0x1980 [f2fs]
[ 330.032004] RSP <ffff88009f44bc40>
[ 330.071560] ---[ end trace 22b76b4a512d7db2 ]---
Cheers
Stephen
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Kernel panic in F2FS mount on NVMe SSD.
2016-05-04 22:47 Kernel panic in F2FS mount on NVMe SSD Stephen Bates
@ 2016-05-05 0:09 ` Jaegeuk Kim
2016-05-05 2:35 ` Stephen Bates
0 siblings, 1 reply; 9+ messages in thread
From: Jaegeuk Kim @ 2016-05-05 0:09 UTC (permalink / raw)
To: Stephen Bates; +Cc: linux-f2fs-devel
Hi Stephen,
On Wed, May 04, 2016 at 10:47:10PM +0000, Stephen Bates wrote:
> Hi
>
> I have been experimenting with putting F2FS onto a NVMe block device. I have been hitting a kernel panic on mount and was wondering if anyone on the list had any ideas what might be causing this. I formatted the NVMe block device using the f2fs-tools version 1.2.0 with default format settings. This is Ubuntu 14.04 running v2.6-rc6 (though I see a similar problem on older kernels).
I think f2fs-tools 1.2.0 is pretty old, since there've made a lot of bug fixes
till 1.6.1 so far.
One suspicious thing like this.
commit 092e3d9da37796daf439e94141c57886d6fc6e50 "mkfs.f2fs: large volume support"
So could you build the latest one from here?
http://git.kernel.org/cgit/linux/kernel/git/jaegeuk/f2fs-tools.git
And, do you mean kernel v4.6-rc6?
Thanks,
>
> The NVMe device is about 4TB big and is formatted for 4KB sectors. I have done some long term stability testing on the same drive using ext4 so the issue does seem to be F2FS related.
>
> [ 328.464442]
> [ 328.464449] **********************************************************
> [ 328.464451] ** NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE **
> [ 328.464453] ** **
> [ 328.464455] ** trace_printk() being used. Allocating extra memory. **
> [ 328.464456] ** **
> [ 328.464458] ** This means that this is a DEBUG kernel and it is **
> [ 328.464460] ** unsafe for production use. **
> [ 328.464461] ** **
> [ 328.464463] ** If you see this message and you are not debugging **
> [ 328.464465] ** the kernel, report this immediately to your vendor! **
> [ 328.464466] ** **
> [ 328.464468] ** NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE **
> [ 328.464470] **********************************************************
> [ 330.031867] ------------[ cut here ]------------
> [ 330.031962] kernel BUG at fs/f2fs/segment.h:608!
> [ 330.032004] invalid opcode: 0000 [#1] SMP
> [ 330.032004] Modules linked in: crc32_generic(E) f2fs(E) xt_tcpudp(E) ip6table_filter(E) ip6_tables(E) iptable_filter(E) ip_tables(E) x_tables(E) des_generic(E) md4(E) rfcomm(E) bnep(E) bluetooth(E) nls_utf8(E) cifs(E) fscache(E) amdkfd(E) amd_iommu_v2(E) radeon(E) i2c_algo_bit(E) drm_kms_helper(E) syscopyarea(E) sysfillrect(E) joydev(E) sysimgblt(E) fb_sys_fops(E) ttm(E) drm(E) nvme(E) gpio_ich(E) nvme_core(E) serio_raw(E) coretemp(E) i5000_edac(E) 8250_fintek(E) edac_core(E) lpc_ich(E) shpchp(E) mac_hid(E) i5k_amb(E) parport_pc(E) ppdev(E) lp(E) parport(E) hid_generic(E) usbhid(E) hid(E) pata_acpi(E) e1000e(E) ptp(E) pps_core(E) floppy(E)
> [ 330.032004] CPU: 1 PID: 2913 Comm: mount Tainted: G E 4.6.0-rc6+nevado-v4.6-rc6 #2
> [ 330.032004] Hardware name: Supermicro X7DB8/X7DB8, BIOS 2.1 06/23/2008
> [ 330.032004] task: ffff8800cabe3400 ti: ffff88009f448000 task.ti: ffff88009f448000
> [ 330.032004] RIP: 0010:[<ffffffffc0509e1f>] [<ffffffffc0509e1f>] build_segment_manager+0x191f/0x1980 [f2fs]
> [ 330.032004] RSP: 0018:ffff88009f44bc40 EFLAGS: 00010297
> [ 330.032004] RAX: 0000000000000000 RBX: 00000000001a30a1 RCX: 00000000000001fe
> [ 330.032004] RDX: 0000000000000000 RSI: 0000000000000200 RDI: ffff88009f44bc80
> [ 330.032004] RBP: ffff88009f44bcf8 R08: 3ffffffffe9da846 R09: ffffffffffffffff
> [ 330.032004] R10: ffff88010ceeecb8 R11: ffff88009f44bb60 R12: ffffc90008177338
> [ 330.032004] R13: 0000000000000072 R14: 0000000000000200 R15: ffff8800c15b1800
> [ 330.032004] FS: 00007f3728b35880(0000) GS:ffff88012fc80000(0000) knlGS:0000000000000000
> [ 330.032004] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 330.032004] CR2: 00007f9530eecf70 CR3: 00000000a2e23000 CR4: 00000000000006e0
> [ 330.032004] Stack:
> [ 330.032004] 0000000000008137 00007c0000000400 ffff8800001aa400 ffff8801292b0f00
> [ 330.032004] ffff8800001a30a1 ffff8800cabb0120 ffff8800b98dd400 8000000000000000
> [ 330.032004] 4000000000f7874d 3000000000432464 d000000002c71521 9000000000550b07
> [ 330.032004] Call Trace:
> [ 330.032004] [<ffffffffc04f2bbd>] f2fs_fill_super+0x62d/0xfe0 [f2fs]
> [ 330.032004] [<ffffffff81209217>] mount_bdev+0x187/0x1c0
> [ 330.032004] [<ffffffffc04f2590>] ? f2fs_commit_super+0xc0/0xc0 [f2fs]
> [ 330.032004] [<ffffffffc04ecbe5>] f2fs_mount+0x15/0x20 [f2fs]
> [ 330.032004] [<ffffffff81209af9>] mount_fs+0x39/0x160
> [ 330.032004] [<ffffffff811a6145>] ? __alloc_percpu+0x15/0x20
> [ 330.032004] [<ffffffff81224c47>] vfs_kern_mount+0x67/0x110
> [ 330.032004] [<ffffffff812271f5>] do_mount+0x225/0xdb0
> [ 330.032004] [<ffffffff811e542f>] ? __kmalloc_track_caller+0x1af/0x250
> [ 330.032004] [<ffffffff811a0a31>] ? strndup_user+0x41/0x80
> [ 330.032004] [<ffffffff811a0952>] ? memdup_user+0x42/0x70
> [ 330.032004] [<ffffffff81228063>] SyS_mount+0x83/0xd0
> [ 330.032004] [<ffffffff817aecf6>] entry_SYSCALL_64_fastpath+0x1e/0xa8
> [ 330.032004] Code: 0e 48 83 c7 04 48 83 c6 04 83 ea 04 89 4f fc e9 8d ed ff ff 48 83 ea 01 e9 7a ef ff ff 48 8d 7a ff e9 d6 ee ff ff e8 3d 61 00 00 <0f> 0b 4c 8d 68 ff e9 af f7 ff ff 48 83 e8 01 e9 87 f7 ff ff e8
> [ 330.032004] RIP [<ffffffffc0509e1f>] build_segment_manager+0x191f/0x1980 [f2fs]
> [ 330.032004] RSP <ffff88009f44bc40>
> [ 330.071560] ---[ end trace 22b76b4a512d7db2 ]---
>
> Cheers
>
> Stephen
>
>
> ------------------------------------------------------------------------------
> Find and fix application performance issues faster with Applications Manager
> Applications Manager provides deep performance insights into multiple tiers of
> your business applications. It resolves application problems quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> _______________________________________________
> Linux-f2fs-devel mailing list
> Linux-f2fs-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Kernel panic in F2FS mount on NVMe SSD.
2016-05-05 0:09 ` Jaegeuk Kim
@ 2016-05-05 2:35 ` Stephen Bates
2016-05-05 3:48 ` Jaegeuk Kim
0 siblings, 1 reply; 9+ messages in thread
From: Stephen Bates @ 2016-05-05 2:35 UTC (permalink / raw)
To: Jaegeuk Kim; +Cc: linux-f2fs-devel
> Hi Stephen,
>
> On Wed, May 04, 2016 at 10:47:10PM +0000, Stephen Bates wrote:
> > Hi
> >
> > I have been experimenting with putting F2FS onto a NVMe block device. I
> have been hitting a kernel panic on mount and was wondering if anyone on
> the list had any ideas what might be causing this. I formatted the NVMe block
> device using the f2fs-tools version 1.2.0 with default format settings. This is
> Ubuntu 14.04 running v2.6-rc6 (though I see a similar problem on older
> kernels).
>
> I think f2fs-tools 1.2.0 is pretty old, since there've made a lot of bug fixes till
> 1.6.1 so far.
>
> One suspicious thing like this.
>
> commit 092e3d9da37796daf439e94141c57886d6fc6e50 "mkfs.f2fs: large
> volume support"
>
> So could you build the latest one from here?
> http://git.kernel.org/cgit/linux/kernel/git/jaegeuk/f2fs-tools.git
Yes that fixed the mount panic issue, thanks!
>
> And, do you mean kernel v4.6-rc6?
Yes ;-).
>
> Thanks,
>
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/expected for F2FS? Could it be because I have the IO_TRACE and/or _SECURITY set in .config?
CONFIG_F2FS_FS=m
CONFIG_F2FS_STAT_FS=y
CONFIG_F2FS_FS_XATTR=y
CONFIG_F2FS_FS_POSIX_ACL=y
CONFIG_F2FS_FS_SECURITY=y
CONFIG_F2FS_CHECK_FS=y
# CONFIG_F2FS_FS_ENCRYPTION is not set
CONFIG_F2FS_IO_TRACE=y
Stephen
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Kernel panic in F2FS mount on NVMe SSD.
2016-05-05 2:35 ` Stephen Bates
@ 2016-05-05 3:48 ` Jaegeuk Kim
2016-05-05 17:53 ` Stephen Bates
0 siblings, 1 reply; 9+ messages in thread
From: Jaegeuk Kim @ 2016-05-05 3:48 UTC (permalink / raw)
To: Stephen Bates; +Cc: linux-f2fs-devel
On Thu, May 05, 2016 at 02:35:42AM +0000, Stephen Bates wrote:
> > Hi Stephen,
> >
> > On Wed, May 04, 2016 at 10:47:10PM +0000, Stephen Bates wrote:
> > > Hi
> > >
> > > I have been experimenting with putting F2FS onto a NVMe block device. I
> > have been hitting a kernel panic on mount and was wondering if anyone on
> > the list had any ideas what might be causing this. I formatted the NVMe block
> > device using the f2fs-tools version 1.2.0 with default format settings. This is
> > Ubuntu 14.04 running v2.6-rc6 (though I see a similar problem on older
> > kernels).
> >
> > I think f2fs-tools 1.2.0 is pretty old, since there've made a lot of bug fixes till
> > 1.6.1 so far.
> >
> > One suspicious thing like this.
> >
> > commit 092e3d9da37796daf439e94141c57886d6fc6e50 "mkfs.f2fs: large
> > volume support"
> >
> > So could you build the latest one from here?
> > http://git.kernel.org/cgit/linux/kernel/git/jaegeuk/f2fs-tools.git
>
> Yes that fixed the mount panic issue, thanks!
>
> >
> > And, do you mean kernel v4.6-rc6?
>
> Yes ;-).
>
> >
> > Thanks,
> >
>
> 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/expected for F2FS? Could it be because I have the IO_TRACE and/or _SECURITY set in .config?
Which fallocate did you test among expand, punch_hole, or something else?
Let me check that especially.
Actually, I have not tested its speed considerably, but more focused on
functionality and stability.
If you're trying to measure FS performance, please turn off:
- most of Lock Debugging configs (e.g., mutex, spinlock and rw-lock)
- CONFIG_F2FS_CHECK_FS
- CONFIG_F2FS_IO_TRACE
Thanks,
>
> CONFIG_F2FS_FS=m
> CONFIG_F2FS_STAT_FS=y
> CONFIG_F2FS_FS_XATTR=y
> CONFIG_F2FS_FS_POSIX_ACL=y
> CONFIG_F2FS_FS_SECURITY=y
> CONFIG_F2FS_CHECK_FS=y
> # CONFIG_F2FS_FS_ENCRYPTION is not set
> CONFIG_F2FS_IO_TRACE=y
>
> Stephen
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Kernel panic in F2FS mount on NVMe SSD.
2016-05-05 3:48 ` Jaegeuk Kim
@ 2016-05-05 17:53 ` Stephen Bates
2016-05-07 17:49 ` Jaegeuk Kim
0 siblings, 1 reply; 9+ messages in thread
From: Stephen Bates @ 2016-05-05 17:53 UTC (permalink / raw)
To: Jaegeuk Kim; +Cc: linux-f2fs-devel
> >
> > 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/expected for F2FS? Could it be
> because I have the IO_TRACE and/or _SECURITY set in .config?
>
> Which fallocate did you test among expand, punch_hole, or something else?
> Let me check that especially.
> Actually, I have not tested its speed considerably, but more focused on
> functionality and stability.
I just issued a
fallocate -l 3TiB /mnt/f2fs/tst
right after the mkfs.f2fs call. I am not sure what fallocate() path that excites in the FS. I also (for fun) timed the removal of that file via the rm command and it is also very slow. iostat implies the drive is being read at about 4MB/s while this removal is being performed.
$ sudo time rm /mnt/f2fs/tst
0.00user 307.62system 6:01.75elapsed 85%CPU (0avgtext+0avgdata 1852maxresident)k
2978552inputs+6332168outputs (1major+80minor)pagefaults 0swaps
>
> If you're trying to measure FS performance, please turn off:
> - most of Lock Debugging configs (e.g., mutex, spinlock and rw-lock)
> - CONFIG_F2FS_CHECK_FS
> - CONFIG_F2FS_IO_TRACE
OK. I will try some more testing with those config options...
Stephen
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Kernel panic in F2FS mount on NVMe SSD.
2016-05-05 17:53 ` Stephen Bates
@ 2016-05-07 17:49 ` Jaegeuk Kim
2016-05-09 15:35 ` Stephen Bates
0 siblings, 1 reply; 9+ messages in thread
From: Jaegeuk Kim @ 2016-05-07 17:49 UTC (permalink / raw)
To: Stephen Bates; +Cc: linux-f2fs-devel
On Thu, May 05, 2016 at 05:53:14PM +0000, 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/expected for F2FS? Could it be
> > because I have the IO_TRACE and/or _SECURITY set in .config?
> >
> > Which fallocate did you test among expand, punch_hole, or something else?
> > Let me check that especially.
> > Actually, I have not tested its speed considerably, but more focused on
> > functionality and stability.
>
> I just issued a
>
> fallocate -l 3TiB /mnt/f2fs/tst
>
> right after the mkfs.f2fs call. I am not sure what fallocate() path that excites in the FS. I also (for fun) timed the removal of that file via the rm command and it is also very slow. iostat implies the drive is being read at about 4MB/s while this removal is being performed.
>
> $ sudo time rm /mnt/f2fs/tst
> 0.00user 307.62system 6:01.75elapsed 85%CPU (0avgtext+0avgdata 1852maxresident)k
> 2978552inputs+6332168outputs (1major+80minor)pagefaults 0swaps
We could figure out how to improve them. :)
I've been preparing two patches under testing right now.
- f2fs: fallocate data blocks in single locked node page
- f2fs: read node blocks ahead when truncating blocks
Thanks,
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Kernel panic in F2FS mount on NVMe SSD.
2016-05-07 17:49 ` Jaegeuk Kim
@ 2016-05-09 15:35 ` Stephen Bates
2016-05-09 17:26 ` Jaegeuk Kim
0 siblings, 1 reply; 9+ messages in thread
From: Stephen Bates @ 2016-05-09 15:35 UTC (permalink / raw)
To: Jaegeuk Kim; +Cc: linux-f2fs-devel
>
> We could figure out how to improve them. :)
>
> I've been preparing two patches under testing right now.
> - f2fs: fallocate data blocks in single locked node page
> - f2fs: read node blocks ahead when truncating blocks
>
> Thanks,
Let me know when and where I can pull these and I can do some testing too.
Thanks!
Stephen
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Kernel panic in F2FS mount on NVMe SSD.
2016-05-09 15:35 ` Stephen Bates
@ 2016-05-09 17:26 ` Jaegeuk Kim
2016-05-14 0:59 ` Jaegeuk Kim
0 siblings, 1 reply; 9+ messages in thread
From: Jaegeuk Kim @ 2016-05-09 17:26 UTC (permalink / raw)
To: Stephen Bates; +Cc: linux-f2fs-devel
On Mon, May 09, 2016 at 03:35:25PM +0000, Stephen Bates wrote:
> >
> > We could figure out how to improve them. :)
> >
> > I've been preparing two patches under testing right now.
> > - f2fs: fallocate data blocks in single locked node page
> > - f2fs: read node blocks ahead when truncating blocks
> >
> > Thanks,
>
> Let me know when and where I can pull these and I can do some testing too.
Okay, Chao Yu gave more improvement, so now I'm starting to do stress tests.
Once getting some good signs, I'll let you know.
Meanwhile, you may take a look at any testing patches from:
http://git.kernel.org/cgit/linux/kernel/git/jaegeuk/f2fs.git/log/?h=dev
Thanks,
>
> Thanks!
>
> Stephen
------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: Kernel panic in F2FS mount on NVMe SSD.
2016-05-09 17:26 ` Jaegeuk Kim
@ 2016-05-14 0:59 ` Jaegeuk Kim
0 siblings, 0 replies; 9+ messages in thread
From: Jaegeuk Kim @ 2016-05-14 0:59 UTC (permalink / raw)
To: Stephen Bates; +Cc: linux-f2fs-devel
Hello,
On Mon, May 09, 2016 at 10:26:08AM -0700, Jaegeuk Kim wrote:
> On Mon, May 09, 2016 at 03:35:25PM +0000, Stephen Bates wrote:
> > >
> > > We could figure out how to improve them. :)
> > >
> > > I've been preparing two patches under testing right now.
> > > - f2fs: fallocate data blocks in single locked node page
> > > - f2fs: read node blocks ahead when truncating blocks
> > >
> > > Thanks,
> >
> > Let me know when and where I can pull these and I can do some testing too.
>
> Okay, Chao Yu gave more improvement, so now I'm starting to do stress tests.
> Once getting some good signs, I'll let you know.
>
> Meanwhile, you may take a look at any testing patches from:
> http://git.kernel.org/cgit/linux/kernel/git/jaegeuk/f2fs.git/log/?h=dev
I think the patches are ready.
You may be able to test with the above up-to-date branch.
Thanks,
>
> Thanks,
>
> >
> > Thanks!
> >
> > Stephen
>
> ------------------------------------------------------------------------------
> Find and fix application performance issues faster with Applications Manager
> Applications Manager provides deep performance insights into multiple tiers of
> your business applications. It resolves application problems quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> _______________________________________________
> Linux-f2fs-devel mailing list
> Linux-f2fs-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
------------------------------------------------------------------------------
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2016-05-14 0:59 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-05-04 22:47 Kernel panic in F2FS mount on NVMe SSD Stephen Bates
2016-05-05 0:09 ` Jaegeuk Kim
2016-05-05 2:35 ` Stephen Bates
2016-05-05 3:48 ` Jaegeuk Kim
2016-05-05 17:53 ` Stephen Bates
2016-05-07 17:49 ` Jaegeuk Kim
2016-05-09 15:35 ` Stephen Bates
2016-05-09 17:26 ` Jaegeuk Kim
2016-05-14 0:59 ` Jaegeuk Kim
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.