Age | Commit message (Collapse) | Author |
|
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull vfs updates from Al Viro:
"Assorted cleanups and fixes.
Probably the most interesting part long-term is ->d_init() - that will
have a bunch of followups in (at least) ceph and lustre, but we'll
need to sort the barrier-related rules before it can get used for
really non-trivial stuff.
Another fun thing is the merge of ->d_iput() callers (dentry_iput()
and dentry_unlink_inode()) and a bunch of ->d_compare() ones (all
except the one in __d_lookup_lru())"
* 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (26 commits)
fs/dcache.c: avoid soft-lockup in dput()
vfs: new d_init method
vfs: Update lookup_dcache() comment
bdev: get rid of ->bd_inodes
Remove last traces of ->sync_page
new helper: d_same_name()
dentry_cmp(): use lockless_dereference() instead of smp_read_barrier_depends()
vfs: clean up documentation
vfs: document ->d_real()
vfs: merge .d_select_inode() into .d_real()
unify dentry_iput() and dentry_unlink_inode()
binfmt_misc: ->s_root is not going anywhere
drop redundant ->owner initializations
ufs: get rid of redundant checks
orangefs: constify inode_operations
missed comment updates from ->direct_IO() prototype change
file_inode(f)->i_mapping is f->f_mapping
trim fsnotify hooks a bit
9p: new helper - v9fs_parent_fid()
debugfs: ->d_parent is never NULL or negative
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip
Pull xen updates from David Vrabel:
"Features and fixes for 4.8-rc0:
- ACPI support for guests on ARM platforms.
- Generic steal time support for arm and x86.
- Support cases where kernel cpu is not Xen VCPU number (e.g., if
in-guest kexec is used).
- Use the system workqueue instead of a custom workqueue in various
places"
* tag 'for-linus-4.8-rc0-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip: (47 commits)
xen: add static initialization of steal_clock op to xen_time_ops
xen/pvhvm: run xen_vcpu_setup() for the boot CPU
xen/evtchn: use xen_vcpu_id mapping
xen/events: fifo: use xen_vcpu_id mapping
xen/events: use xen_vcpu_id mapping in events_base
x86/xen: use xen_vcpu_id mapping when pointing vcpu_info to shared_info
x86/xen: use xen_vcpu_id mapping for HYPERVISOR_vcpu_op
xen: introduce xen_vcpu_id mapping
x86/acpi: store ACPI ids from MADT for future usage
x86/xen: update cpuid.h from Xen-4.7
xen/evtchn: add IOCTL_EVTCHN_RESTRICT
xen-blkback: really don't leak mode property
xen-blkback: constify instance of "struct attribute_group"
xen-blkfront: prefer xenbus_scanf() over xenbus_gather()
xen-blkback: prefer xenbus_scanf() over xenbus_gather()
xen: support runqueue steal time on xen
arm/xen: add support for vm_assist hypercall
xen: update xen headers
xen-pciback: drop superfluous variables
xen-pciback: short-circuit read path used for merging write values
...
|
|
Merge updates from Andrew Morton:
- a few misc bits
- ocfs2
- most(?) of MM
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (125 commits)
thp: fix comments of __pmd_trans_huge_lock()
cgroup: remove unnecessary 0 check from css_from_id()
cgroup: fix idr leak for the first cgroup root
mm: memcontrol: fix documentation for compound parameter
mm: memcontrol: remove BUG_ON in uncharge_list
mm: fix build warnings in <linux/compaction.h>
mm, thp: convert from optimistic swapin collapsing to conservative
mm, thp: fix comment inconsistency for swapin readahead functions
thp: update Documentation/{vm/transhuge,filesystems/proc}.txt
shmem: split huge pages beyond i_size under memory pressure
thp: introduce CONFIG_TRANSPARENT_HUGE_PAGECACHE
khugepaged: add support of collapse for tmpfs/shmem pages
shmem: make shmem_inode_info::lock irq-safe
khugepaged: move up_read(mmap_sem) out of khugepaged_alloc_page()
thp: extract khugepaged from mm/huge_memory.c
shmem, thp: respect MADV_{NO,}HUGEPAGE for file mappings
shmem: add huge pages support
shmem: get_unmapped_area align huge page
shmem: prepare huge= mount option and sysfs knob
mm, rmap: account shmem thp pages
...
|
|
Zsmalloc is ready for page migration so zram can use __GFP_MOVABLE from
now on.
I did test to see how it helps to make higher order pages. Test
scenario is as follows.
KVM guest, 1G memory, ext4 formated zram block device,
for i in `seq 1 8`;
do
dd if=/dev/vda1 of=mnt/test$i.txt bs=128M count=1 &
done
wait `pidof dd`
for i in `seq 1 2 8`;
do
rm -rf mnt/test$i.txt
done
fstrim -v mnt
echo "init"
cat /proc/buddyinfo
echo "compaction"
echo 1 > /proc/sys/vm/compact_memory
cat /proc/buddyinfo
old:
init
Node 0, zone DMA 208 120 51 41 11 0 0 0 0 0 0
Node 0, zone DMA32 16380 13777 9184 3805 789 54 3 0 0 0 0
compaction
Node 0, zone DMA 132 82 40 39 16 2 1 0 0 0 0
Node 0, zone DMA32 5219 5526 4969 3455 1831 677 139 15 0 0 0
new:
init
Node 0, zone DMA 379 115 97 19 2 0 0 0 0 0 0
Node 0, zone DMA32 18891 16774 10862 3947 637 21 0 0 0 0 0
compaction
Node 0, zone DMA 214 66 87 29 10 3 0 0 0 0 0
Node 0, zone DMA32 1612 3139 3154 2469 1745 990 384 94 7 0 0
As you can see, compaction made so many high-order pages. Yay!
Link: http://lkml.kernel.org/r/1464736881-24886-13-git-send-email-minchan@kernel.org
Signed-off-by: Minchan Kim <minchan@kernel.org>
Reviewed-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
We now allocate streams from CPU_UP hot-plug path, there are no
context-dependent stream allocations anymore and we can schedule from
zcomp_strm_alloc(). Use GFP_KERNEL directly and drop a gfp_t parameter.
Link: http://lkml.kernel.org/r/20160531122017.2878-9-sergey.senozhatsky@gmail.com
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Add "deflate", "lz4hc", "842" algorithms to the list of known
compression backends. The real availability of those algorithms,
however, depends on the corresponding CONFIG_CRYPTO_FOO config options.
[sergey.senozhatsky@gmail.com: zram-add-more-compression-algorithms-v3]
Link: http://lkml.kernel.org/r/20160604024902.11778-7-sergey.senozhatsky@gmail.com
Link: http://lkml.kernel.org/r/20160531122017.2878-8-sergey.senozhatsky@gmail.com
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Remove lzo/lz4 backends, we use crypto API now.
[sergey.senozhatsky@gmail.com: zram-delete-custom-lzo-lz4-v3]
Link: http://lkml.kernel.org/r/20160604024902.11778-6-sergey.senozhatsky@gmail.com
Link: http://lkml.kernel.org/r/20160531122017.2878-7-sergey.senozhatsky@gmail.com
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
There is no way to get a string with all the crypto comp algorithms
supported by the crypto comp engine, so we need to maintain our own
backends list. At the same time we additionally need to use
crypto_has_comp() to make sure that the user has requested a compression
algorithm that is recognized by the crypto comp engine. Relying on
/proc/crypto is not an options here, because it does not show
not-yet-inserted compression modules.
Example:
modprobe zram
cat /proc/crypto | grep -i lz4
modprobe lz4
cat /proc/crypto | grep -i lz4
name : lz4
driver : lz4-generic
module : lz4
So the user can't tell exactly if the lz4 is really supported from
/proc/crypto output, unless someone or something has loaded it.
This patch also adds crypto_has_comp() to zcomp_available_show(). We
store all the compression algorithms names in zcomp's `backends' array,
regardless the CONFIG_CRYPTO_FOO configuration, but show only those that
are also supported by crypto engine. This helps user to know the exact
list of compression algorithms that can be used.
Example:
module lz4 is not loaded yet, but is supported by the crypto
engine. /proc/crypto has no information on this module, while
zram's `comp_algorithm' lists it:
cat /proc/crypto | grep -i lz4
cat /sys/block/zram0/comp_algorithm
[lzo] lz4 deflate lz4hc 842
We still use the `backends' array to determine if the requested
compression backend is known to crypto api. This array, however, may not
contain some entries, therefore as the last step we call crypto_has_comp()
function which attempts to insmod the requested compression algorithm to
determine if crypto api supports it. The advantage of this method is that
now we permit the usage of out-of-tree crypto compression modules
(implementing S/W or H/W compression).
[sergey.senozhatsky@gmail.com: zram-use-crypto-api-to-check-alg-availability-v3]
Link: http://lkml.kernel.org/r/20160604024902.11778-4-sergey.senozhatsky@gmail.com
Link: http://lkml.kernel.org/r/20160531122017.2878-5-sergey.senozhatsky@gmail.com
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
We don't have an idle zstreams list anymore and our write path now works
absolutely differently, preventing preemption during compression. This
removes possibilities of read paths preempting writes at wrong places
(which could badly affect the performance of both paths) and at the same
time opens the door for a move from custom LZO/LZ4 compression backends
implementation to a more generic one, using crypto compress API.
Joonsoo Kim [1] attempted to do this a while ago, but faced with the
need of introducing a new crypto API interface. The root cause was the
fact that crypto API compression algorithms require a compression stream
structure (in zram terminology) for both compression and decompression
ops, while in reality only several of compression algorithms really need
it. This resulted in a concept of context-less crypto API compression
backends [2]. Both write and read paths, though, would have been
executed with the preemption enabled, which in the worst case could have
resulted in a decreased worst-case performance, e.g. consider the
following case:
CPU0
zram_write()
spin_lock()
take the last idle stream
spin_unlock()
<< preempted >>
zram_read()
spin_lock()
no idle streams
spin_unlock()
schedule()
resuming zram_write compression()
but it took me some time to realize that, and it took even longer to
evolve zram and to make it ready for crypto API. The key turned out to be
-- drop the idle streams list entirely. Without the idle streams list we
are free to use compression algorithms that require compression stream for
decompression (read), because streams are now placed in per-cpu data and
each write path has to disable preemption for compression op, almost
completely eliminating the aforementioned case (technically, we still have
a small chance, because write path has a fast and a slow paths and the
slow path is executed with the preemption enabled; but the frequency of
failed fast path is too low).
TEST
====
- 4 CPUs, x86_64 system
- 3G zram, lzo
- fio tests: read, randread, write, randwrite, rw, randrw
test script [3] command:
ZRAM_SIZE=3G LOG_SUFFIX=XXXX FIO_LOOPS=5 ./zram-fio-test.sh
BASE PATCHED
jobs1
READ: 2527.2MB/s 2482.7MB/s
READ: 2102.7MB/s 2045.0MB/s
WRITE: 1284.3MB/s 1324.3MB/s
WRITE: 1080.7MB/s 1101.9MB/s
READ: 430125KB/s 437498KB/s
WRITE: 430538KB/s 437919KB/s
READ: 399593KB/s 403987KB/s
WRITE: 399910KB/s 404308KB/s
jobs2
READ: 8133.5MB/s 7854.8MB/s
READ: 7086.6MB/s 6912.8MB/s
WRITE: 3177.2MB/s 3298.3MB/s
WRITE: 2810.2MB/s 2871.4MB/s
READ: 1017.6MB/s 1023.4MB/s
WRITE: 1018.2MB/s 1023.1MB/s
READ: 977836KB/s 984205KB/s
WRITE: 979435KB/s 985814KB/s
jobs3
READ: 13557MB/s 13391MB/s
READ: 11876MB/s 11752MB/s
WRITE: 4641.5MB/s 4682.1MB/s
WRITE: 4164.9MB/s 4179.3MB/s
READ: 1453.8MB/s 1455.1MB/s
WRITE: 1455.1MB/s 1458.2MB/s
READ: 1387.7MB/s 1395.7MB/s
WRITE: 1386.1MB/s 1394.9MB/s
jobs4
READ: 20271MB/s 20078MB/s
READ: 18033MB/s 17928MB/s
WRITE: 6176.8MB/s 6180.5MB/s
WRITE: 5686.3MB/s 5705.3MB/s
READ: 2009.4MB/s 2006.7MB/s
WRITE: 2007.5MB/s 2004.9MB/s
READ: 1929.7MB/s 1935.6MB/s
WRITE: 1926.8MB/s 1932.6MB/s
jobs5
READ: 18823MB/s 19024MB/s
READ: 18968MB/s 19071MB/s
WRITE: 6191.6MB/s 6372.1MB/s
WRITE: 5818.7MB/s 5787.1MB/s
READ: 2011.7MB/s 1981.3MB/s
WRITE: 2011.4MB/s 1980.1MB/s
READ: 1949.3MB/s 1935.7MB/s
WRITE: 1940.4MB/s 1926.1MB/s
jobs6
READ: 21870MB/s 21715MB/s
READ: 19957MB/s 19879MB/s
WRITE: 6528.4MB/s 6537.6MB/s
WRITE: 6098.9MB/s 6073.6MB/s
READ: 2048.6MB/s 2049.9MB/s
WRITE: 2041.7MB/s 2042.9MB/s
READ: 2013.4MB/s 1990.4MB/s
WRITE: 2009.4MB/s 1986.5MB/s
jobs7
READ: 21359MB/s 21124MB/s
READ: 19746MB/s 19293MB/s
WRITE: 6660.4MB/s 6518.8MB/s
WRITE: 6211.6MB/s 6193.1MB/s
READ: 2089.7MB/s 2080.6MB/s
WRITE: 2085.8MB/s 2076.5MB/s
READ: 2041.2MB/s 2052.5MB/s
WRITE: 2037.5MB/s 2048.8MB/s
jobs8
READ: 20477MB/s 19974MB/s
READ: 18922MB/s 18576MB/s
WRITE: 6851.9MB/s 6788.3MB/s
WRITE: 6407.7MB/s 6347.5MB/s
READ: 2134.8MB/s 2136.1MB/s
WRITE: 2132.8MB/s 2134.4MB/s
READ: 2074.2MB/s 2069.6MB/s
WRITE: 2087.3MB/s 2082.4MB/s
jobs9
READ: 19797MB/s 19994MB/s
READ: 18806MB/s 18581MB/s
WRITE: 6878.7MB/s 6822.7MB/s
WRITE: 6456.8MB/s 6447.2MB/s
READ: 2141.1MB/s 2154.7MB/s
WRITE: 2144.4MB/s 2157.3MB/s
READ: 2084.1MB/s 2085.1MB/s
WRITE: 2091.5MB/s 2092.5MB/s
jobs10
READ: 19794MB/s 19784MB/s
READ: 18794MB/s 18745MB/s
WRITE: 6984.4MB/s 6676.3MB/s
WRITE: 6532.3MB/s 6342.7MB/s
READ: 2150.6MB/s 2155.4MB/s
WRITE: 2156.8MB/s 2161.5MB/s
READ: 2106.4MB/s 2095.6MB/s
WRITE: 2109.7MB/s 2098.4MB/s
BASE PATCHED
jobs1 perfstat
stalled-cycles-frontend 102,480,595,419 ( 41.53%) 114,508,864,804 ( 46.92%)
stalled-cycles-backend 51,941,417,832 ( 21.05%) 46,836,112,388 ( 19.19%)
instructions 283,612,054,215 ( 1.15) 283,918,134,959 ( 1.16)
branches 56,372,560,385 ( 724.923) 56,449,814,753 ( 733.766)
branch-misses 374,826,000 ( 0.66%) 326,935,859 ( 0.58%)
jobs2 perfstat
stalled-cycles-frontend 155,142,745,777 ( 40.99%) 164,170,979,198 ( 43.82%)
stalled-cycles-backend 70,813,866,387 ( 18.71%) 66,456,858,165 ( 17.74%)
instructions 463,436,648,173 ( 1.22) 464,221,890,191 ( 1.24)
branches 91,088,733,902 ( 760.088) 91,278,144,546 ( 769.133)
branch-misses 504,460,363 ( 0.55%) 394,033,842 ( 0.43%)
jobs3 perfstat
stalled-cycles-frontend 201,300,397,212 ( 39.84%) 223,969,902,257 ( 44.44%)
stalled-cycles-backend 87,712,593,974 ( 17.36%) 81,618,888,712 ( 16.19%)
instructions 642,869,545,023 ( 1.27) 644,677,354,132 ( 1.28)
branches 125,724,560,594 ( 690.682) 126,133,159,521 ( 694.542)
branch-misses 527,941,798 ( 0.42%) 444,782,220 ( 0.35%)
jobs4 perfstat
stalled-cycles-frontend 246,701,197,429 ( 38.12%) 280,076,030,886 ( 43.29%)
stalled-cycles-backend 119,050,341,112 ( 18.40%) 110,955,641,671 ( 17.15%)
instructions 822,716,962,127 ( 1.27) 825,536,969,320 ( 1.28)
branches 160,590,028,545 ( 688.614) 161,152,996,915 ( 691.068)
branch-misses 650,295,287 ( 0.40%) 550,229,113 ( 0.34%)
jobs5 perfstat
stalled-cycles-frontend 298,958,462,516 ( 38.30%) 344,852,200,358 ( 44.16%)
stalled-cycles-backend 137,558,742,122 ( 17.62%) 129,465,067,102 ( 16.58%)
instructions 1,005,714,688,752 ( 1.29) 1,007,657,999,432 ( 1.29)
branches 195,988,773,962 ( 697.730) 196,446,873,984 ( 700.319)
branch-misses 695,818,940 ( 0.36%) 624,823,263 ( 0.32%)
jobs6 perfstat
stalled-cycles-frontend 334,497,602,856 ( 36.71%) 387,590,419,779 ( 42.38%)
stalled-cycles-backend 163,539,365,335 ( 17.95%) 152,640,193,639 ( 16.69%)
instructions 1,184,738,177,851 ( 1.30) 1,187,396,281,677 ( 1.30)
branches 230,592,915,640 ( 702.902) 231,253,802,882 ( 702.356)
branch-misses 747,934,786 ( 0.32%) 643,902,424 ( 0.28%)
jobs7 perfstat
stalled-cycles-frontend 396,724,684,187 ( 37.71%) 460,705,858,952 ( 43.84%)
stalled-cycles-backend 188,096,616,496 ( 17.88%) 175,785,787,036 ( 16.73%)
instructions 1,364,041,136,608 ( 1.30) 1,366,689,075,112 ( 1.30)
branches 265,253,096,936 ( 700.078) 265,890,524,883 ( 702.839)
branch-misses 784,991,589 ( 0.30%) 729,196,689 ( 0.27%)
jobs8 perfstat
stalled-cycles-frontend 440,248,299,870 ( 36.92%) 509,554,793,816 ( 42.46%)
stalled-cycles-backend 222,575,930,616 ( 18.67%) 213,401,248,432 ( 17.78%)
instructions 1,542,262,045,114 ( 1.29) 1,545,233,932,257 ( 1.29)
branches 299,775,178,439 ( 697.666) 300,528,458,505 ( 694.769)
branch-misses 847,496,084 ( 0.28%) 748,794,308 ( 0.25%)
jobs9 perfstat
stalled-cycles-frontend 506,269,882,480 ( 37.86%) 592,798,032,820 ( 44.43%)
stalled-cycles-backend 253,192,498,861 ( 18.93%) 233,727,666,185 ( 17.52%)
instructions 1,721,985,080,913 ( 1.29) 1,724,666,236,005 ( 1.29)
branches 334,517,360,255 ( 694.134) 335,199,758,164 ( 697.131)
branch-misses 873,496,730 ( 0.26%) 815,379,236 ( 0.24%)
jobs10 perfstat
stalled-cycles-frontend 549,063,363,749 ( 37.18%) 651,302,376,662 ( 43.61%)
stalled-cycles-backend 281,680,986,810 ( 19.07%) 277,005,235,582 ( 18.55%)
instructions 1,901,859,271,180 ( 1.29) 1,906,311,064,230 ( 1.28)
branches 369,398,536,153 ( 694.004) 370,527,696,358 ( 688.409)
branch-misses 967,929,335 ( 0.26%) 890,125,056 ( 0.24%)
BASE PATCHED
seconds elapsed 79.421641008 78.735285546
seconds elapsed 61.471246133 60.869085949
seconds elapsed 62.317058173 62.224188495
seconds elapsed 60.030739363 60.081102518
seconds elapsed 74.070398362 74.317582865
seconds elapsed 84.985953007 85.414364176
seconds elapsed 97.724553255 98.173311344
seconds elapsed 109.488066758 110.268399318
seconds elapsed 122.768189405 122.967164498
seconds elapsed 135.130035105 136.934770801
On my other system (8 x86_64 CPUs, short version of test results):
BASE PATCHED
seconds elapsed 19.518065994 19.806320662
seconds elapsed 15.172772749 15.594718291
seconds elapsed 13.820925970 13.821708564
seconds elapsed 13.293097816 14.585206405
seconds elapsed 16.207284118 16.064431606
seconds elapsed 17.958376158 17.771825767
seconds elapsed 19.478009164 19.602961508
seconds elapsed 21.347152811 21.352318709
seconds elapsed 24.478121126 24.171088735
seconds elapsed 26.865057442 26.767327618
So performance-wise the numbers are quite similar.
Also update zcomp interface to be more aligned with the crypto API.
[1] http://marc.info/?l=linux-kernel&m=144480832108927&w=2
[2] http://marc.info/?l=linux-kernel&m=145379613507518&w=2
[3] https://github.com/sergey-senozhatsky/zram-perf-test
Link: http://lkml.kernel.org/r/20160531122017.2878-3-sergey.senozhatsky@gmail.com
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Suggested-by: Minchan Kim <minchan@kernel.org>
Suggested-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
This has started as a 'add zlib support' work, but after some thinking I
saw no blockers for a bigger change -- a switch to crypto API.
We don't have an idle zstreams list anymore and our write path now works
absolutely differently, preventing preemption during compression. This
removes possibilities of read paths preempting writes at wrong places
and opens the door for a move from custom LZO/LZ4 compression backends
implementation to a more generic one, using crypto compress API.
This patch set also eliminates the need of a new context-less crypto API
interface, which was quite hard to sell, so we can move along faster.
benchmarks:
(x86_64, 4GB, zram-perf script)
perf reported run-time fio (max jobs=3). I performed fio test with the
increasing number of parallel jobs (max to 3) on a 3G zram device, using
`static' data and the following crypto comp algorithms:
842, deflate, lz4, lz4hc, lzo
the output was:
- test running time (which can tell us what algorithms performs faster)
and
- zram mm_stat (which tells the compressed memory size, max used memory, etc).
It's just for information. for example, LZ4HC has twice the running
time of LZO, but the compressed memory size is: 23592960 vs 34603008
bytes.
test-fio-zram-842
197.907655282 seconds time elapsed
201.623142884 seconds time elapsed
226.854291345 seconds time elapsed
test-fio-zram-DEFLATE
253.259516155 seconds time elapsed
258.148563401 seconds time elapsed
290.251909365 seconds time elapsed
test-fio-zram-LZ4
27.022598717 seconds time elapsed
29.580522717 seconds time elapsed
33.293463430 seconds time elapsed
test-fio-zram-LZ4HC
56.393954615 seconds time elapsed
74.904659747 seconds time elapsed
101.940998564 seconds time elapsed
test-fio-zram-LZO
28.155948075 seconds time elapsed
30.390036330 seconds time elapsed
34.455773159 seconds time elapsed
zram mm_stat-s (max fio jobs=3)
test-fio-zram-842
mm_stat (jobs1): 3221225472 673185792 690266112 0 690266112 0 0
mm_stat (jobs2): 3221225472 673185792 690266112 0 690266112 0 0
mm_stat (jobs3): 3221225472 673185792 690266112 0 690266112 0 0
test-fio-zram-DEFLATE
mm_stat (jobs1): 3221225472 24379392 37761024 0 37761024 0 0
mm_stat (jobs2): 3221225472 24379392 37761024 0 37761024 0 0
mm_stat (jobs3): 3221225472 24379392 37761024 0 37761024 0 0
test-fio-zram-LZ4
mm_stat (jobs1): 3221225472 23592960 37761024 0 37761024 0 0
mm_stat (jobs2): 3221225472 23592960 37761024 0 37761024 0 0
mm_stat (jobs3): 3221225472 23592960 37761024 0 37761024 0 0
test-fio-zram-LZ4HC
mm_stat (jobs1): 3221225472 23592960 37761024 0 37761024 0 0
mm_stat (jobs2): 3221225472 23592960 37761024 0 37761024 0 0
mm_stat (jobs3): 3221225472 23592960 37761024 0 37761024 0 0
test-fio-zram-LZO
mm_stat (jobs1): 3221225472 34603008 50335744 0 50335744 0 0
mm_stat (jobs2): 3221225472 34603008 50335744 0 50335744 0 0
mm_stat (jobs3): 3221225472 34603008 50335744 0 50339840 0 0
This patch (of 8):
We don't perform any zstream idle list lookup anymore, so
zcomp_strm_find()/zcomp_strm_release() names are not representative.
Rename to zcomp_stream_get()/zcomp_stream_put().
Link: http://lkml.kernel.org/r/20160531122017.2878-2-sergey.senozhatsky@gmail.com
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Pull block driver updates from Jens Axboe:
"This branch also contains core changes. I've come to the conclusion
that from 4.9 and forward, I'll be doing just a single branch. We
often have dependencies between core and drivers, and it's hard to
always split them up appropriately without pulling core into drivers
when that happens.
That said, this contains:
- separate secure erase type for the core block layer, from
Christoph.
- set of discard fixes, from Christoph.
- bio shrinking fixes from Christoph, as a followup up to the
op/flags change in the core branch.
- map and append request fixes from Christoph.
- NVMeF (NVMe over Fabrics) code from Christoph. This is pretty
exciting!
- nvme-loop fixes from Arnd.
- removal of ->driverfs_dev from Dan, after providing a
device_add_disk() helper.
- bcache fixes from Bhaktipriya and Yijing.
- cdrom subchannel read fix from Vchannaiah.
- set of lightnvm updates from Wenwei, Matias, Johannes, and Javier.
- set of drbd updates and fixes from Fabian, Lars, and Philipp.
- mg_disk error path fix from Bart.
- user notification for failed device add for loop, from Minfei.
- NVMe in general:
+ NVMe delay quirk from Guilherme.
+ SR-IOV support and command retry limits from Keith.
+ fix for memory-less NUMA node from Masayoshi.
+ use UINT_MAX for discard sectors, from Minfei.
+ cancel IO fixes from Ming.
+ don't allocate unused major, from Neil.
+ error code fixup from Dan.
+ use constants for PSDT/FUSE from James.
+ variable init fix from Jay.
+ fabrics fixes from Ming, Sagi, and Wei.
+ various fixes"
* 'for-4.8/drivers' of git://git.kernel.dk/linux-block: (115 commits)
nvme/pci: Provide SR-IOV support
nvme: initialize variable before logical OR'ing it
block: unexport various bio mapping helpers
scsi/osd: open code blk_make_request
target: stop using blk_make_request
block: simplify and export blk_rq_append_bio
block: ensure bios return from blk_get_request are properly initialized
virtio_blk: use blk_rq_map_kern
memstick: don't allow REQ_TYPE_BLOCK_PC requests
block: shrink bio size again
block: simplify and cleanup bvec pool handling
block: get rid of bio_rw and READA
block: don't ignore -EOPNOTSUPP blkdev_issue_write_same
block: introduce BLKDEV_DISCARD_ZERO to fix zeroout
NVMe: don't allocate unused nvme_major
nvme: avoid crashes when node 0 is memoryless node.
nvme: Limit command retries
loop: Make user notify for adding loop device failed
nvme-loop: fix nvme-loop Kconfig dependencies
nvmet: fix return value check in nvmet_subsys_alloc()
...
|
|
Pull core block updates from Jens Axboe:
- the big change is the cleanup from Mike Christie, cleaning up our
uses of command types and modified flags. This is what will throw
some merge conflicts
- regression fix for the above for btrfs, from Vincent
- following up to the above, better packing of struct request from
Christoph
- a 2038 fix for blktrace from Arnd
- a few trivial/spelling fixes from Bart Van Assche
- a front merge check fix from Damien, which could cause issues on
SMR drives
- Atari partition fix from Gabriel
- convert cfq to highres timers, since jiffies isn't granular enough
for some devices these days. From Jan and Jeff
- CFQ priority boost fix idle classes, from me
- cleanup series from Ming, improving our bio/bvec iteration
- a direct issue fix for blk-mq from Omar
- fix for plug merging not involving the IO scheduler, like we do for
other types of merges. From Tahsin
- expose DAX type internally and through sysfs. From Toshi and Yigal
* 'for-4.8/core' of git://git.kernel.dk/linux-block: (76 commits)
block: Fix front merge check
block: do not merge requests without consulting with io scheduler
block: Fix spelling in a source code comment
block: expose QUEUE_FLAG_DAX in sysfs
block: add QUEUE_FLAG_DAX for devices to advertise their DAX support
Btrfs: fix comparison in __btrfs_map_block()
block: atari: Return early for unsupported sector size
Doc: block: Fix a typo in queue-sysfs.txt
cfq-iosched: Charge at least 1 jiffie instead of 1 ns
cfq-iosched: Fix regression in bonnie++ rewrite performance
cfq-iosched: Convert slice_resid from u64 to s64
block: Convert fifo_time from ulong to u64
blktrace: avoid using timespec
block/blk-cgroup.c: Declare local symbols static
block/bio-integrity.c: Add #include "blk.h"
block/partition-generic.c: Remove a set-but-not-used variable
block: bio: kill BIO_MAX_SIZE
cfq-iosched: temporarily boost queue priority for idle classes
block: drbd: avoid to use BIO_MAX_SIZE
block: bio: remove BIO_MAX_SECTORS
...
|
|
Commit 9d092603cc ("xen-blkback: do not leak mode property") left one
path unfixed; correct this.
Acked-by: Jens Axboe <axboe@kernel.dk>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
|
|
The functions these get passed to have been taking pointers to const
since at least 2.6.16.
Acked-by: Jens Axboe <axboe@kernel.dk>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
|
|
... for single items being collected: It is more typesafe (as the
compiler can check format string and to-be-written-to variable match)
and requires one less parameter to be passed.
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
|
|
... for single items being collected: It is more typesafe (as the
compiler can check format string and to-be-written-to variable match)
and requires one less parameter to be passed.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Jens Axboe <axboe@kernel.dk>
|
|
Currently, presence of direct_access() in block_device_operations
indicates support of DAX on its block device. Because
block_device_operations is instantiated with 'const', this DAX
capablity may not be enabled conditinally.
In preparation for supporting DAX to device-mapper devices, add
QUEUE_FLAG_DAX to request_queue flags to advertise their DAX
support. This will allow to set the DAX capability based on how
mapped device is composed.
Signed-off-by: Toshi Kani <toshi.kani@hpe.com>
Acked-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: <linux-s390@vger.kernel.org>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
blk_get_request is used for BLOCK_PC and similar passthrough requests.
Currently we always need to call blk_rq_set_block_pc or an open coded
version of it to allow appending bios using the request mapping helpers
later on, which is a somewhat awkward API. Instead move the
initialization part of blk_rq_set_block_pc into blk_get_request, so that
we always have a safe to use request.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Similar to how SCSI and NVMe prepare passthrough requests. This avoids
poking into request internals too much.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
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: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
There is no error number returned if loop driver fails in function
alloc_disk to add new loop device. Add a correct error number to make
user notify in this case.
Signed-off-by: Minfei Huang <mnghuan@gmail.com>
Reviewed-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm into for-4.8/drivers
Dan writes:
"The removal of ->driverfs_dev in favor of just passing the parent
device in as a parameter to add_disk(). See below, it has received a
"Reviewed-by" from Christoph, Bart, and Johannes.
It is also a pre-requisite for Fam Zheng's work to cleanup gendisk
uevents vs attribute visibility [1]. We would extend device_add_disk()
to take an attribute_group list.
This is based off a branch of block.git/for-4.8/drivers and has
received a positive build success notification from the kbuild robot
across several configs.
[1]: "gendisk: Generate uevent after attribute available"
http://marc.info/?l=linux-virtualization&m=146725201522201&w=2"
|
|
Pull block IO fixes from Jens Axboe:
"Three small fixes that have been queued up and tested for this series:
- A bug fix for xen-blkfront from Bob Liu, fixing an issue with
incomplete requests during migration.
- A fix for an ancient issue in retrieving the IO priority of a
different PID than self, preventing that task from going away while
we access it. From Omar.
- A writeback fix from Tahsin, fixing a case where we'd call ihold()
with a zero ref count inode"
* 'for-linus' of git://git.kernel.dk/linux-block:
block: fix use-after-free in sys_ioprio_get()
writeback: inode cgroup wb switch should not call ihold()
xen-blkfront: save uncompleted reqs in blkfront_resume()
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs into work.misc
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen into for-linus
|
|
Uncompleted reqs used to be 'saved and resubmitted' in blkfront_recover() during
migration, but that's too late after multi-queue was introduced.
After a migrate to another host (which may not have multiqueue support), the
number of rings (block hardware queues) may be changed and the ring and shadow
structure will also be reallocated.
The blkfront_recover() then can't 'save and resubmit' the real
uncompleted reqs because shadow structure have been reallocated.
This patch fixes this issue by moving the 'save' logic out of
blkfront_recover() to earlier place in blkfront_resume().
The 'resubmit' is not changed and still in blkfront_recover().
Signed-off-by: Bob Liu <bob.liu@oracle.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: stable@vger.kernel.org
|
|
MG_DISK_MAJ is defined as 0 so dynamic block major number
allocation is used by the driver and the assigned major
number is stored in host->major. This patch fixes error
path in mg_probe() to use host->major instead of using
MG_DISK_MAJ.
Cc: unsik Kim <donari75@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
For block drivers that specify a parent device, convert them to use
device_add_disk().
This conversion was done with the following semantic patch:
@@
struct gendisk *disk;
expression E;
@@
- disk->driverfs_dev = E;
...
- add_disk(disk);
+ device_add_disk(E, disk);
@@
struct gendisk *disk;
expression E1, E2;
@@
- disk->driverfs_dev = E1;
...
E2 = disk;
...
- add_disk(E2);
+ device_add_disk(E1, E2);
...plus some manual fixups for a few missed conversions.
Cc: Jens Axboe <axboe@fb.com>
Cc: Keith Busch <keith.busch@intel.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: James Bottomley <James.Bottomley@hansenpartnership.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
This is the third version of the patchset previously sent [1]. I have
basically only rebased it on top of 4.7-rc1 tree and dropped "dm: get
rid of superfluous gfp flags" which went through dm tree. I am sending
it now because it is tree wide and chances for conflicts are reduced
considerably when we want to target rc2. I plan to send the next step
and rename the flag and move to a better semantic later during this
release cycle so we will have a new semantic ready for 4.8 merge window
hopefully.
Motivation:
While working on something unrelated I've checked the current usage of
__GFP_REPEAT in the tree. It seems that a majority of the usage is and
always has been bogus because __GFP_REPEAT has always been about costly
high order allocations while we are using it for order-0 or very small
orders very often. It seems that a big pile of them is just a
copy&paste when a code has been adopted from one arch to another.
I think it makes some sense to get rid of them because they are just
making the semantic more unclear. Please note that GFP_REPEAT is
documented as
* __GFP_REPEAT: Try hard to allocate the memory, but the allocation attempt
* _might_ fail. This depends upon the particular VM implementation.
while !costly requests have basically nofail semantic. So one could
reasonably expect that order-0 request with __GFP_REPEAT will not loop
for ever. This is not implemented right now though.
I would like to move on with __GFP_REPEAT and define a better semantic
for it.
$ git grep __GFP_REPEAT origin/master | wc -l
111
$ git grep __GFP_REPEAT | wc -l
36
So we are down to the third after this patch series. The remaining
places really seem to be relying on __GFP_REPEAT due to large allocation
requests. This still needs some double checking which I will do later
after all the simple ones are sorted out.
I am touching a lot of arch specific code here and I hope I got it right
but as a matter of fact I even didn't compile test for some archs as I
do not have cross compiler for them. Patches should be quite trivial to
review for stupid compile mistakes though. The tricky parts are usually
hidden by macro definitions and thats where I would appreciate help from
arch maintainers.
[1] http://lkml.kernel.org/r/1461849846-27209-1-git-send-email-mhocko@kernel.org
This patch (of 19):
__GFP_REPEAT has a rather weak semantic but since it has been introduced
around 2.6.12 it has been ignored for low order allocations. Yet we
have the full kernel tree with its usage for apparently order-0
allocations. This is really confusing because __GFP_REPEAT is
explicitly documented to allow allocation failures which is a weaker
semantic than the current order-0 has (basically nofail).
Let's simply drop __GFP_REPEAT from those places. This would allow to
identify place which really need allocator to retry harder and formulate
a more specific semantic for what the flag is supposed to do actually.
Link: http://lkml.kernel.org/r/1464599699-30131-2-git-send-email-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: "James E.J. Bottomley" <jejb@parisc-linux.org>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Chen Liqin <liqin.linux@gmail.com>
Cc: Chris Metcalf <cmetcalf@mellanox.com> [for tile]
Cc: Guan Xuetao <gxt@mprc.pku.edu.cn>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Helge Deller <deller@gmx.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Cc: John Crispin <blogic@openwrt.org>
Cc: Lennox Wu <lennox.wu@gmail.com>
Cc: Ley Foon Tan <lftan@altera.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Rich Felker <dalias@libc.org>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
crypto_alloc_hash returns an ERR_PTR(), not NULL.
Also reset peer_integrity_tfm to NULL, to not call crypto_free_hash()
on an errno in the cleanup path.
Reported-by: Insu Yun <wuninsu@gmail.com>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
For larger devices, the array of bitmap page pointers can grow very
large (8000 pointers per TB of storage).
For each activity log transaction, we need to flush the associated
bitmap pages to stable storage. Currently, we just "mark" the respective
pages while setting up the transaction, then tell the bitmap code to
write out all marked pages, but skip unchanged pages.
But one such transaction can affect only a small number of bitmap pages,
there is no need to scan the full array of several (ten-)thousand
page pointers to find the few marked ones.
Instead, remember the index numbers of the few affected pages,
and later only re-check those to skip duplicates and unchanged ones.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Also skip the message unless bitmap IO took longer than 5 ms.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
This should silence a warning about an empty statement. Thanks to Fabian
Frederick <fabf@skynet.be> who sent a patch I modified to be smaller and
avoids an additional indent level.
Signed-off-by: Roland Kammerer <roland.kammerer@linbit.com>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
This contains various cosmetic fixes ranging from simple typos to
const-ifying, and using booleans properly.
Original commit messages from Fabian's patch set:
drbd: debugfs: constify drbd_version_fops
drbd: use seq_put instead of seq_print where possible
drbd: include linux/uaccess.h instead of asm/uaccess.h
drbd: use const char * const for drbd strings
drbd: kerneldoc warning fix in w_e_end_data_req()
drbd: use unsigned for one bit fields
drbd: use bool for peer is_ states
drbd: fix typo
drbd: use | for bitmask combination
drbd: use true/false for bool
drbd: fix drbd_bm_init() comments
drbd: introduce peer state union
drbd: fix maybe_pull_ahead() locking comments
drbd: use bool for growing
drbd: remove redundant declarations
drbd: replace if/BUG by BUG_ON
Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: Roland Kammerer <roland.kammerer@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Scenario, starting with normal operation
Connected Primary/Secondary UpToDate/UpToDate
NetworkFailure Primary/Unknown UpToDate/DUnknown (frozen)
... more failures happen, secondary loses it's disk,
but eventually is able to re-establish the replication link ...
Connected Primary/Secondary UpToDate/Diskless (resumed; needs to bump uuid!)
We used to just resume/resent suspended requests,
without bumping the UUID.
Which will lead to problems later, when we want to re-attach the disk on
the peer, without first disconnecting, or if we experience additional
failures, because we now have diverging data without being able to
recognize it.
Make sure we also bump the current data generation UUID,
if we notice "peer disk unknown" -> "peer disk known bad".
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
We already serialize connection state changes,
and other, non-connection state changes (role changes)
while we are establishing a connection.
But if we have an established connection,
then trigger a resync handshake (by primary --force or similar),
until now we just had to be "lucky".
Consider this sequence (e.g. deployment scenario):
create-md; up;
-> Connected Secondary/Secondary Inconsistent/Inconsistent
then do a racy primary --force on both peers.
block drbd0: drbd_sync_handshake:
block drbd0: self 0000000000000004:0000000000000000:0000000000000000:0000000000000000 bits:25590 flags:0
block drbd0: peer 0000000000000004:0000000000000000:0000000000000000:0000000000000000 bits:25590 flags:0
block drbd0: peer( Unknown -> Secondary ) conn( WFReportParams -> Connected ) pdsk( DUnknown -> Inconsistent )
block drbd0: peer( Secondary -> Primary ) pdsk( Inconsistent -> UpToDate )
*** HERE things go wrong. ***
block drbd0: role( Secondary -> Primary )
block drbd0: drbd_sync_handshake:
block drbd0: self 0000000000000005:0000000000000000:0000000000000000:0000000000000000 bits:25590 flags:0
block drbd0: peer C90D2FC716D232AB:0000000000000004:0000000000000000:0000000000000000 bits:25590 flags:0
block drbd0: Becoming sync target due to disk states.
block drbd0: Writing the whole bitmap, full sync required after drbd_sync_handshake.
block drbd0: Remote failed to finish a request within 6007ms > ko-count (2) * timeout (30 * 0.1s)
drbd s0: peer( Primary -> Unknown ) conn( Connected -> Timeout ) pdsk( UpToDate -> DUnknown )
The problem here is that the local promotion happens before the sync handshake
triggered by the remote promotion was completed. Some assumptions elsewhere
become wrong, and when the expected resync handshake is then received and
processed, we get stuck in a deadlock, which can only be recovered by reboot :-(
Fix: if we know the peer has good data,
and our own disk is present, but NOT good,
and there is no resync going on yet,
we expect a sync handshake to happen "soon".
So reject a racy promotion with SS_IN_TRANSIENT_STATE.
Result:
... as above ...
block drbd0: peer( Secondary -> Primary ) pdsk( Inconsistent -> UpToDate )
*** local promotion being postponed until ... ***
block drbd0: drbd_sync_handshake:
block drbd0: self 0000000000000004:0000000000000000:0000000000000000:0000000000000000 bits:25590 flags:0
block drbd0: peer 77868BDA836E12A5:0000000000000004:0000000000000000:0000000000000000 bits:25590 flags:0
...
block drbd0: conn( WFBitMapT -> WFSyncUUID )
block drbd0: updated sync uuid 85D06D0E8887AD44:0000000000000000:0000000000000000:0000000000000000
block drbd0: conn( WFSyncUUID -> SyncTarget )
*** ... after the resync handshake ***
block drbd0: role( Secondary -> Primary )
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
If in a two-primary scenario, we lost our peer, freeze IO,
and are still frozen (no UUID rotation) when the peer comes back
as Secondary after a hard crash, we will see identical UUIDs.
The "rule_nr = 40" chose to use the "CRASHED_PRIMARY" bit as
arbitration, but that would cause the still running (but frozen) Primary
to become SyncTarget (which it typically refuses), and the handshake is
declined.
Fix: check current roles.
If we have *one* current primary, the Primary wins.
(rule_nr = 41)
Since that is a protocol change, use the newly introduced DRBD_FF_WSAME
to determine if rule_nr = 41 can be applied.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
We will support WRITE_SAME, if
* all peers support WRITE_SAME (both in kernel and DRBD version),
* all peer devices support WRITE_SAME
* logical_block_size is identical on all peers.
We may at some point introduce a fallback on the receiving side
for devices/kernels that do not support WRITE_SAME,
by open-coding a submit loop. But not yet.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Even if discard_zeroes_data != 0,
if discard_zeroes_if_aligned is set, we assume we can reliably
zero-out/discard using the drbd_issue_peer_discard() helper.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
When re-attaching the local backend device to a C_STANDALONE D_DISKLESS
R_PRIMARY with OND_SUSPEND_IO, we may only resume IO if we recognize the
backend that is being attached as D_UP_TO_DATE.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
If DRBD lost all path to good data,
and the on-no-data-accessible policy is OND_SUSPEND_IO,
all pending and new IO requests are suspended (will block).
If that setting is OND_IO_ERROR, IO will still be completed.
READ to "clean" areas (e.g. on an D_INCONSISTENT device,
and bitmap indicates a block is already in sync) will succeed.
READ to "unclean" areas (bitmap indicates block is out-of-sync),
will return EIO.
If we are already D_DISKLESS (or D_FAILED), we also return EIO.
Unfortunately, on a former R_PRIMARY C_SYNC_TARGET D_INCONSISTENT,
after replication link loss, new WRITE requests still went through OK.
The would also set the "out-of-sync" bit on their way, so READ after
WRITE would still return EIO. Also, the data generation UUIDs had not
been bumped, we would cause data divergence, without being able to
detect it on the next sync handshake, given the right sequence of events
in a multiple error scenario and "improper" order of recovery actions.
The right thing to do is to return EIO for all new writes,
unless we have access to good, current, D_UP_TO_DATE data.
The "established best practices" way to avoid these situations in the
first place is to set OND_SUSPEND_IO, or even do a hard-reset from
the pri-on-incon-degr policy helper hook.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Possibly sequence of events:
SyncTarget is made Primary, then loses replication link
(only path to good data on SyncSource).
Behavior is then controlled by the on-no-data-accessible policy,
which defaults to OND_IO_ERROR (may be set to OND_SUSPEND_IO).
If OND_IO_ERROR is in fact the current policy, we clear the susp_fen
(IO suspended due to fencing policy) flag, do NOT set the susp_nod
(IO suspended due to no data) flag.
But we forgot to call the IO error completion for all pending,
suspended, requests.
While at it, also add a race check for a theoretically possible
race with a new handshake (network hickup), we may be able to
re-send requests, and can avoid passing IO errors up the stack.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
When resync is finished, we already call the "after-resync-target"
handler (on the former sync target, obviously), once per volume.
Paired with the before-resync-target handler, you can create snapshots,
before the resync causes the volumes to become inconsistent,
and discard those snapshots again, once they are no longer needed.
It was also overloaded to be paired with the "fence-peer" handler,
to "unfence" once the volumes are up-to-date and known good.
This has some disadvantages, though: we call "fence-peer" for the whole
connection (once for the group of volumes), but would call unfence as
side-effect of after-resync-target once for each volume.
Also, we fence on a (current, or about to become) Primary,
which will later become the sync-source.
Calling unfence only as a side effect of the after-resync-target
handler opens a race window, between a new fence on the Primary
(SyncTarget) and the unfence on the SyncTarget, which is difficult to
close without some kind of "cluster wide lock" in those handlers.
We would not need those handlers if we could still communicate.
Which makes trying to aquire a cluster wide lock from those handlers
seem like a very bad idea.
This introduces the "unfence-peer" handler, which will be called
per connection (once for the group of volumes), just like the fence
handler, only once all volumes are back in sync, and on the SyncSource.
Which is expected to be the node that previously called "fence", the
node that is currently allowed to be Primary, and thus the only node
that could trigger a new "fence" that could race with this unfence.
Which makes us not need any cluster wide synchronization here,
serializing two scripts running on the same node is trivial.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
If the replication link breaks exactly during "resync finished" detection,
finishing too early on the sync source could again lead to UUIDs rotated
too fast, and potentially a spurious full resync on next handshake.
Always wait for explicit resync finished state change notification from
the sync target.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Make sure we have at least 67 (> AL_UPDATES_PER_TRANSACTION)
al-extents available, and allow up to half of that to be
discarded in one bio.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
For consistency, also zero-out partial unaligned chunks of discard
requests on the local backend.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Now that we have the discard_zeroes_if_aligned setting, we should also
check it when setting up our queue parameters on the primary,
not only on the receiving side.
We announce discard support,
UNLESS
* we are connected to a peer that does not support TRIM
on the DRBD protocol level. Otherwise, it would either discard, or
do a fallback to zero-out, depending on its backend and configuration.
* our local backend does not support discards,
or (discard_zeroes_data=0 AND discard_zeroes_if_aligned=no).
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
We can avoid spurious data divergence caused by partially-ignored
discards on certain backends with discard_zeroes_data=0, if we
translate partial unaligned discard requests into explicit zero-out.
The relevant use case is LVM/DM thin.
If on different nodes, DRBD is backed by devices with differing
discard characteristics, discards may lead to data divergence
(old data or garbage left over on one backend, zeroes due to
unmapped areas on the other backend). Online verify would now
potentially report tons of spurious differences.
While probably harmless for most use cases (fstrim on a file system),
DRBD cannot have that, it would violate our promise to upper layers
that our data instances on the nodes are identical.
To be correct and play safe (make sure data is identical on both copies),
we would have to disable discard support, if our local backend (on a
Primary) does not support "discard_zeroes_data=true".
We'd also have to translate discards to explicit zero-out on the
receiving (typically: Secondary) side, unless the receiving side
supports "discard_zeroes_data=true".
Which both would allocate those blocks, instead of unmapping them,
in contrast with expectations.
LVM/DM thin does set discard_zeroes_data=0,
because it silently ignores discards to partial chunks.
We can work around this by checking the alignment first.
For unaligned (wrt. alignment and granularity) or too small discards,
we zero-out the initial (and/or) trailing unaligned partial chunks,
but discard all the aligned full chunks.
At least for LVM/DM thin, the result is effectively "discard_zeroes_data=1".
Arguably it should behave this way internally, by default,
and we'll try to make that happen.
But our workaround is still valid for already deployed setups,
and for other devices that may behave this way.
Setting discard-zeroes-if-aligned=yes will allow DRBD to use
discards, and to announce discard_zeroes_data=true, even on
backends that announce discard_zeroes_data=false.
Setting discard-zeroes-if-aligned=no will cause DRBD to always
fall-back to zero-out on the receiving side, and to not even
announce discard capabilities on the Primary, if the respective
backend announces discard_zeroes_data=false.
We used to ignore the discard_zeroes_data setting completely.
To not break established and expected behaviour, and suddenly
cause fstrim on thin-provisioned LVs to run out-of-space,
instead of freeing up space, the default value is "yes".
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
To maintain write-order fidelity accros all volumes in a DRBD resource,
the receiver of a P_BARRIER needs to issue flushes to all volumes.
We used to do this by calling blkdev_issue_flush(), synchronously,
one volume at a time.
We now submit all flushes to all volumes in parallel, then wait for all
completions, to reduce worst-case latencies on multi-volume resources.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|