Update the stathz description to reflect reality. profhz is the only
thing we should deprecate. Add some implementation notes that describe
the optimizations made to date.
Discusssed with: emaste
Reviewed by: kib (prior), jhb (prior), gbe
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D30815
This field is used only by illumos mdb. On other platforms it only
increases the struct size from 32 to 40 bytes. For struct vdev_queue
including 13 instances of avl_tree_t size means active cache lines.
Keep the padding in user-space for now to not break the ABI.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closes#12290
With default dbuf cache size of 1/32 of ARC, it makes no sense to have
hash table of the same size (or even bigger on Linux). Reduce it to
1/8 of ARC's one, still leaving some slack, assuming higher I/O rate
via dbuf cache than via ARC.
Remove padding from ARC hash locks array. The idea behind padding
is to avoid false sharing between locks. It would have sense if
there would be a limited number of very busy locks. But since we
have no limit on the number, using the same memory for more locks we
can achieve even lower lock contention with the same false sharing,
or we can use less memory for the same contention level.
Reduce number of hash locks from 8192 to 2048. The number is still
big enough to not cause contention, but reduced memory size improves
cache hit rate for mutex_tryenter() in ARC eviction thread, saving
about 1% of the thread time.
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closes#12289
Fix a leak of abd_t that manifested mostly when using
raidzN with at least as many columns as N (e.g. a
four-disk raidz2 but not a three-disk raidz2).
Sufficiently heavy raidz use would eventually run a system
out of memory.
Additionally:
* Switch abd_cache arena to FIRSTFIT, which empirically
improves perofrmance.
* Make abd_chunk_cache more performant and debuggable.
* Allocate the abd_zero_buf from abd_chunk_cache rather
than the heap.
* Don't try to reap non-existent qcaches in abd_cache arena.
* KM_PUSHPAGE->KM_SLEEP when allocating chunks from their
own arena
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Jorgen Lundman <lundman@lundman.net>
Co-authored-by: Sean Doran <smd@use.net>
Closes#12295
dmu_zfetch_stream_fini() is missing calls to destroy the refcounts,
leaking them and the mutex inside.
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Jorgen Lundman <lundman@lundman.net>
Closes#12294
Use dp_dirty_pertxg[] for txg_kick(), instead of dp_dirty_total in
original code. Extra parameter "txg" is added for txg_kick(), thus it
knows which txg to kick. Also txg_kick() call is moved from
dsl_pool_need_dirty_delay() to dsl_pool_dirty_space() so that we can
know the txg number assigned for txg_kick().
Some unnecessary code regarding dp_dirty_total in txg_sync_thread() is
also cleaned up.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: jxdking <lostking2008@hotmail.com>
Closes#12274
The only reason for spa_config_*() to use refcount instead of simple
non-atomic (thanks to scl_lock) variable for scl_count is tracking,
hard disabled for the last 8 years. Switch to simple int scl_count
reduces the lock hold time by avoiding atomic, plus makes structure
fit into single cache line, reducing the locks contention.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Mark Maybee <mark.maybee@delphix.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closes#12287
This introduces a new, per-process flag, "NO_NEW_PRIVS", which
is inherited, preserved on exec, and cannot be cleared. The flag,
when set, makes subsequent execs ignore any SUID and SGID bits,
instead executing those binaries as if they not set.
The main purpose of the flag is implementation of Linux
PROC_SET_NO_NEW_PRIVS prctl(2), and possibly also unpriviledged
chroot.
Reviewed By: kib
Sponsored By: EPSRC
Differential Revision: https://reviews.freebsd.org/D30939
When the setting of kern.ipc.maxsockbuf is less than what is
desired for I/O based on vfs.maxbcachebuf and vfs.nfs.bufpackets,
a console message of "Consider increasing kern.ipc.maxsockbuf".
is printed.
This patch modifies the message to provide a suggested value
for kern.ipc.maxsockbuf.
Note that the setting is only needed when the NFS rsize/wsize
is set to vfs.maxbcachebuf.
While here, make nfs_bufpackets global, so that it can be used
by a future patch that adds a sysctl to set the NFS server's
maximum I/O size. Also, remove "sizeof(u_int32_t)" from the maximum
packet length, since NFS_MAXXDR is already an "overestimate"
of the actual length.
MFC after: 2 weeks
Implement dumping core for Linux binaries on amd64, for both
32- and 64-bit executables. Some bits are still missing.
This is based on a prototype by chuck@.
Reviewed By: kib
Sponsored By: EPSRC
Differential Revision: https://reviews.freebsd.org/D30019
Similarly to what's been done on arm64 with commit
712c060c94, when executing a binary, if the
entry point is a thumb symbol, then make sure we set the PSL_T flag, otherwise
the CPU will interpret it in ARM mode, and that will likely leads to an
undefined instruction.
PR: 256899
MFC after: 1 week
This allows supported libpmc to query/select from the pmu-events table,
which may have a more complete set of events than what we define
manually. A future update to these definitions should greatly improve
this support. The alias table is empty for now, until this future import
is complete.
Add the Foundation's copyright for recent work on this file.
Reviewed by: ray (slightly earlier version)
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D30603
Make it possible to specify event codes without an offset of
PMC_EV_ARMV8_FIRST, by setting a machine-dependent flag. This is
required to make use of event definitions from pmu-events.
Reviewed by: ray (slightly earlier version)
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D30602
This will be used to detect supported pmu events. The expected format is
the MIDR register with the revision and variant fields masked. See also:
lib/libpmc/pmu-events/arch/arm64/mapfile.csv.
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D30601
Eliminate some unnecessary unlocking and relocking when we have to retry
the operation to avoid deadlock. (All of the other pmap functions that
iterate over a PV list already implemented retries without these same
unlocking and relocking operations.)
Reviewed by: kib, markj
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D30951
Per [(1), i.e. test(1), the string comparison operator should be `=`,
not `==` in sh(1) scripts.
No functional change. FreeBSD's test accepts `==` as an alias for `=`
for compatibility with gnu test's `==` extension.
Reviewed by: imp@, gjb@
Pull Request: https://github.com/freebsd/freebsd-src/pull/485
Per tzsetup(8), /etc/localtime and /var/db/zoneinfo go hand in hand.
Remove extra white spaces as well.
Reviewed by: imp@, phk@ (informal no objection email)
Pull Request: https://github.com/freebsd/freebsd-src/pull/486
This enables ZED to auto-online vdevs that are not wholedisk managed by
ZFS.
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Don Brady <don.brady@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
If the entry point for the binary executed is a thumb 2 entry point, make
sure we set the PSR_T bit, or the CPU will interpret it as arm32 code and
bad things will happen.
PR: 256899
MFC after: 1 week
Remove any attempt to use _arm_memcpy and _arm_bzero. It was used by some
xscale platforms to provide functions to use the DMA engine for big
zeroing/copying work, but those platforms are long gone, and it's unlikely
anything else will use those.
Building and installing architecture-specific man pages only raises a number of
problems:
* The https://www.freebsd.org/cgi/man.cgi is incomplete. As an
example, it does not show results for pae(4). The reason for this is
that the cgi interface runs on FreeBSD amd64.
* In FreeBSD amd64 some manual pages have broken X-refs. See hptrr(4)
for an example.
* Also, we have broken links in our Release Notes. This is a
consequence of the first point. See
https://www.freebsd.org/releases/13.0R/hardware/#proc-i386.
Make MAN_ARCH default to 'all' so we build all the man pages for all the
architectures. The difference in disk space is negligible. Also link
architecture-specific man pages to their own section while keeping their own
namespace.
PR: 212290
Reported by: mj@bsdops.com
Approved by: ceri@, wosch@
MFC after: 4 weeks
Finally, we have the correct function definition for strmode. NetBSD/OpenBSD
did this many years ago. This code is weird sign extension safe.
Reviewed by: imp@
Pull Request: https://github.com/freebsd/freebsd-src/pull/493
Increase the Linux-Maximum version in the META file to 5.13.
All of the required compatibility patches have been merged
and the 5.13 kernel has been officially released.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Commit 6fc3099 broke the quoting when invoking the mail program, revert
that change.
Signed-off-by: Laurențiu Nicola <lnicola@dend.ro>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Portsnap uses patches opportunistically to reduce download bandwidth: It
attempts to fetch patches which could be useful, and then makes use of
whichever patches it actually gets. (This solves the otherwise O(n^2)
issue for the server to build patches between every pair of versions.)
During the process of applying patches, portsnap prints lines of the
form "Skipping XXX-YYY (123 of 4567).\r", where the \r serves to allow
each of these (potentially many) lines to overwrite the previous one
on the console. Unfortunately, XXX and YYY here are SHA256 hashes,
resulting in these lines wrapping on reasonable-width consoles.
Replace the hashes with abbreviations of the form "0123...cdef"
(cutting 64 characters down to 11) in order to keep lines to a
reasonable length.
The rather ugly shell code here is used to avoid forking additional
processes; it would be much cleaner using sed(1), but in my testing
the sed-based alternative increases CPU time consumption by 50%.
Requested by: des
To avoid duplication in the vmstat -m output rename the kmalloc type short
description to 'lkpikmalloc' as the Linux emulation layer historically names
its linux malloc type as 'linux'.
Reviewed by: hselasky, kib, emaste
Differential Revision: https://reviews.freebsd.org/D30928
MFC after: 2 weeks
Instead of p_elf_machine use machine member of the Elf_Brandinfo which is now
cached in the struct proc at p_elf_brandinfo member.
Note to MFC: D30918, KBI
Reviewed by: kib, markj
Differential Revision: https://reviews.freebsd.org/D30926
MFC after: 2 weeks
To allow the ABI to make a dicision based on the Brandinfo add a link
to the Elf_Brandinfo into the struct proc. Add a note that the high 8 bits
of Elf_Brandinfo flags is private to the ABI.
Note to MFC: it breaks KBI.
Reviewed by: kib, markj
Differential Revision: https://reviews.freebsd.org/D30918
MFC after: 2 weeks
LLVM's Polly (ISL to be precise) is unhappy with the loop from
ddt_stat_add():
CC [M] fs/zfs/zfs/ddt.o
../lib/External/isl/isl_schedule_node.c:2470: cannot insert node
between set or sequence node and its filter children
(building with the custom patch which adds Polly support to Kbuild)
The mentioned loop is rather suboptimal. All that we need is to just
treat ddt_stat_t as an array of u64 and perform 1:1 addition or
substraction. This can be done in simpler for-loop with the
determined index and bounds. Compiler will expand d_end - d into
a number of ddt_stat_t fields at compile time.
This prevents Polly from failing on this file.
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Lobakin <alobakin@pm.me>
Closes#12253
