Improve zdb -b performance:
- Reduce gethrtime() call to 1/100th of blkptr's;
- Skip manipulating the size-ordered tree;
- Issue more (10, previously 3) async reads;
- Use lighter weight testing in traverse_visitbp();
Illumos issue:
5243 zdb -b could be much faster
Apply upstream 13597:3eac1e8e0f4c (git: illumos-gate@aa846ad9):
Initialize tqent_flags in the userland taskq implementation. Without
this the assertion of tq->tq_freelist != NULL may fail in taskq_destroy.
The problem is that tqent_flags is never initialized in the userland
implementation while the kernel one does initialize it. Without proper
initialization, the flag may have its lowest bit set, making it treated
as TQENT_FLAG_PREALLOC and never removing taskq_ent_t from tq_freelist.
- Limit ARC for zdb at 256MB. zdb do not typically revisit data
in the ARC.
- Increase default max_inflight from 200 to 1000 (can be overriden
by -I) so we can queue more I/Os when doing scrubbing.
- Print status while loading meataslabs for leak detection.
Illumos issues:
5169 zdb should limit its ARC size
5170 zdb -c should create more scrub i/os by default
5171 zdb should print status while loading metaslabs for leak detection
Iterate through all the children instead of returning error when we hit
the first error. This makes the error message give more information
rather than just the first device that causes problem.
Illumos issue:
5118 When verifying or creating a storage pool, error messages only
show one device
Approved by: re (gjb)
Preserve the errno value of an ioctl before calling free(3). Previously,
errno was very occasionally being clobbered, resulting in a bogus error from
dt_consume() and thus an error from dtrace(1).
Don't try to use the 32-bit drti.o unless the data model is explicitly set
to ILP32. Otherwise dtrace -G will attempt to use it on amd64 if it can't
determine which data model to use, which happens when -64 is omitted and
no object files are provided, e.g. with
# dtrace -G -n BEGIN
This would result in a linker error, but now works properly.
Also remove an unnecessary #ifdef.
Import Illumos changes to address the following Illumos issues:
4976 zfs should only avoid writing to a failing non-redundant
top-level vdev
4978 ztest fails in get_metaslab_refcount()
4979 extend free space histogram to device and pool
4980 metaslabs should have a fragmentation metric
4981 remove fragmented ops vector from block allocator
4982 space_map object should proactively upgrade when feature
is enabled
4984 device selection should use fragmentation metric
Don't report non-native block-size pools under zpool status -x
zpool status -x is used to identify pools that are exhibiting
errors or are otherwise unavailable, therefore non-native
block-size pools shouldn't be reported.
Also update man page to clarify other additional conditions
which won't cause a pool to be displayed under zpool status -x.
Sponsored by: Multiplay
Re-apply r248644. This fixes an annoying problem which caused dtrace -c to
fail to attach to stripped binaries. With the _r_debug_postinit symbol,
dtrace(1) can now set a breakpoint in the victim process after it has
registered its DOF table(s) with the kernel. r_debug_state cannot be used
for this purpose since it is called before DOF is made available, in which
case dtrace(1) cannot create USDT probes before the program begins
execution.
Add a function, memstr, which can be used to convert a buffer of
null-separated strings to a single string. This can be used to print the
full arguments of a process using execsnoop (from the DTrace toolkit) or
with the following one-liner:
dtrace -n 'syscall::execve:return {trace(curpsinfo->pr_psargs);}'
Note that this relies on the process arguments being cached via the struct
proc, which means that it will not work for argvs longer than
kern.ps_arg_cache_limit. However, the following rather non-portable
script can be used to extract any argv at exec time:
fbt::kern_execve:entry
{
printf("%s", memstr(args[1]->begin_argv, ' ',
args[1]->begin_envv - args[1]->begin_argv));
}
The debug.dtrace.memstr_max sysctl limits the maximum argument size to
memstr().
Instead of asserting all zio's be properly aligned, only assert
on the logical ones.
Cap uberblocks at 8k, otherwise with ashift=17, there would be
only one uberblock.
This fixes a problem that zdb would trip assert on pools with
ashift >= 0xe (8k).
While there, also change the code so it only attempt to condense
space map unless the uncondensed size consumes greater than
zfs_metaslab_condense_block_threshold blocks.
Illumos issue:
4958 zdb trips assert on pools with ashift >= 0xe
When our linker merges .SUNW_dof sections from multiple files, it simply
concatenates the DOF tables into one section. Previously, the USDT init
code in drti.o would only look at the first table in the DOF section; with
this change, it iterates over all the tables, passing each DOF table to
the kernel.
PR: 186821
Improve extreme rewind import.
When doing an "extreme rewind" import ("zpool import -XF"), we attempt
to verify all data in the pool, essentially scrubbing the entire pool.
The problem is that spa_load_verify_cb() issues an unbounded number of
concurrent scrub i/os. This can lead to all of memory being used for
these zio's, wedging the system. Like normal scrub, we need to put a
cap on the number of outstanding i/os, and have the traverse thread
block when we reach this cap.
For this purpose the cap can be very large (10,000) to optimize the
elevator algorithm. Three kernel tunables have been added:
vfs.zfs.spa_load_verify_maxinflight
vfs.zfs.spa_load_verify_metadata
vfs.zfs.spa_load_verify_data
The latter two tunables controls whether metadata and/or user data
when doing extreme rewind.
Make 'zpool import -T' imply scrub.
Make zpool import -T <txg> accept hexadecimal values for the txg when
prefixed with 0x.
Skip txg's for which there is no uberblock when doing extreme rewind.
Skip reading all user data twice by skipping prefetches when doing
extreme rewinds as we do not access via the ARC.
Illumos issues:
4970 need controls on i/o issued by zpool import -XF
4971 zpool import -T should accept hex values
4972 zpool import -T implies extreme rewind, and thus a scrub
4973 spa_load_retry retries the same txg
4974 spa_load_verify() reads all data twice
Add a new zfs property, "redundant_metadata" which can have values "all" or
"most". The default will be "all", which is the current behavior. When set
to all, ZFS stores an extra copy of all metadata. If a single on-disk block
is corrupt, at worst a single block of user data (which is recordsize bytes
long) can be lost.
Setting to "most" will cause us to only store 1 copy of level-1 indirect
blocks of user data files. This can improve performance of random writes,
because less metadata has to be written. In practice, at worst about
100 blocks (of recordsize bytes each) of user data can be lost if a single
on-disk block is corrupt.
The exact behavior of which metadata blocks are stored redundantly may change
in future releases.
Illumos issue: 3835 zfs need not store 2 copies of all metadata
2915 DTrace in a zone should see "cpu", "curpsinfo", et al
2916 DTrace in a zone should be able to access fds[]
2917 DTrace in a zone should have limited provider access
4477 DTrace should speak JSON
Add stubs for CTF functions which are not yet implemented.
4474 DTrace Userland CTF Support
4475 DTrace userland Keyword
4476 DTrace tests should be better citizens
4479 pid provider types
4480 dof emulation is missing checks
4471 DTrace count() with histogram
4472 DTrace full width distribution histograms
4473 DTrace frequency trails
Merge from r258379 missed the tests.
4248 dtrace(1M) should never create DOF with empty probes section
4249 Only probes from the first DTrace object file will be included
Illumos Revision: 54a20ab41aadcb81c53e72fc65886e964e9add59
Define the KM_NORMALPRI flag for kmem_alloc(), as it is used in some
upstream DTrace code.
MFC r262330:
1452 DTrace buffer autoscaling should be less violent
illumos/illumos-gate@6fb4854bed
Comment out some pointless device open/close around reading device IDs.
FreeBSD ZFS port unlike OpenSolaris does not use device IDs, and does not
implement respective devid_*() fuctions. It is pointless to open devices
just to close them back immediately.
