All the per-loop `libuv` setup remains in `isc_loop`, but the per-thread
RCU setup is moved to `isc_thread` alongside the other per-thread setup.
This avoids repeating the per-thread setup for `call_rcu()` helpers,
and explains a little better why some parts of the per-thread setup
is missing for `call_rcu()` helpers.
This also removes the per-loop `call_rcu()` helpers as we refactored the
isc__random_initialize() in the previous commit.
Remove the `isc_threadarg_t` and `isc_threadresult_t`
typedefs which were unhelpful disguises for `void *`,
and free the dummy jemalloc allocation sooner.
When liburcu is not installed from a system package, its headers are
not treated as system headers by the compiler, so BIND's -Werror and
other warning options take effect. The liburcu headers have a lot
of inline functions, some of which do not use all their arguments,
which BIND's build treats as an error.
This commit allows BIND 9 to be compiled with different flavours of
Userspace RCU, and improves the integration between Userspace RCU and
our event loop:
- In the RCU QSBR, the thread is put offline when polling and online
when rcu_dereference, rcu_assign_pointer (or friends) are called.
- In other RCU modes, we check that we are not reading when reaching the
quiescent callback in the event loop.
- We register the thread before uv_work_run() callback is called and
after it has finished. The rcu_(un)register_thread() has a large
overhead, but that's fine in this case.
There's a recurring pattern walking the ISC_LISTs that just repeats over
and over. Add two macros:
* ISC_LIST_FOREACH(list, elt, link) - walk the static list
* ISC_LIST_FOREACH_SAFE(list, elt, link, next) - walk the list in
a manner that's safe against list member deletions
The spinlock is small (atomic_uint_fast32_t at most), lightweight
synchronization primitive and should only be used for short-lived and
most of the time a isc_mutex should be used.
Add a isc_spinlock unit which is either (most of the time) a think
wrapper around pthread_spin API or an efficient shim implementation of
the simple spinlock.
When shutting down TCP sockets, the read callback calling logic was
flawed, it would call either one less callback or one extra. Fix the
logic in the way:
1. When isc_nm_read() has been called but isc_nm_read_stop() hasn't on
the handle, the read callback will be called with ISC_R_CANCELED to
cancel active reading from the socket/handle.
2. When isc_nm_read() has been called and isc_nm_read_stop() has been
called on the on the handle, the read callback will be called with
ISC_R_SHUTTINGDOWN to signal that the dormant (not-reading) socket
is being shut down.
3. The .reading and .recv_read flags are little bit tricky. The
.reading flag indicates if the outer layer is reading the data (that
would be uv_tcp_t for TCP and isc_nmsocket_t (TCP) for TLSStream),
the .recv_read flag indicates whether somebody is interested in the
data read from the socket.
Usually, you would expect that the .reading should be false when
.recv_read is false, but it gets even more tricky with TLSStream as
the TLS protocol might need to read from the socket even when sending
data.
Fix the usage of the .recv_read and .reading flags in the TLSStream
to their true meaning - which mostly consist of using .recv_read
everywhere and then wrapping isc_nm_read() and isc_nm_read_stop()
with the .reading flag.
4. The TLS failed read helper has been modified to resemble the TCP code
as much as possible, clearing and re-setting the .recv_read flag in
the TCP timeout code has been fixed and .recv_read is now cleared
when isc_nm_read_stop() has been called on the streaming socket.
5. The use of Network Manager in the named_controlconf, isccc_ccmsg, and
isc_httpd units have been greatly simplified due to the improved design.
6. More unit tests for TCP and TLS testing the shutdown conditions have
been added.
Co-authored-by: Ondřej Surý <ondrej@isc.org>
Co-authored-by: Artem Boldariev <artem@isc.org>
In e185412872, the TCP accept quota code
became broken in a subtle way - the quota would get initialized on the
first accept for the server socket and then deleted from the server
socket, so it would never get applied again.
Properly fixing this required a bigger refactoring of the isc_quota API
code to make it much simpler. The new code decouples the ownership of
the quota and acquiring/releasing the quota limit.
After (during) the refactoring it became more clear that we need to use
the callback from the child side of the accepted connection, and not the
server side.
The isc_tid() function is often called on the hot-path and it's the only
function is to return thread_local variable, make the isc_tid() function
a header-only to save several function calls during query-response
processing.
This commit optimises isc_dnsstream_assembler_t in such a way that
memory copying and reallocation are avoided when receiving one or more
complete DNS messages at once. We try to handle the data from the
messages directly, without storing them in an intermediate memory
buffer.
The `isc_histosummary_t` functions were written in the early days of
`hg64` and carried over when I brought `hg64` into BIND. They were
intended to be useful for graphing cumulative frequency distributions
and the like, but in practice whatever draws charts is better off with
a raw histogram export. Especially because of the poor performance of
the old functions.
The replacement `isc_histo_quantiles()` function is intended for
providing a few quantile values in BIND's stats channel, when the user
does not want the full histogram. Unlike the old functions, the caller
provides all the query fractions up-front, so that the values can be
found in a single scan instead of a scan per value. The scan is from
larger values to smaller, since larger quantiles are usually more
interesting, so the scan can bail out early.
Although an `isc_histo_t` is thread-safe, it can suffer
from cache contention under heavy load. To avoid this,
an `isc_histomulti_t` contains a histogram per thread,
so updates are local and low-contention.
This is an adaptation of my `hg64` experiments for use in BIND.
As well as renaming everything according to ISC style, I have
written some more extensive tests that ensure the edge cases are
correct and the fenceposts are in the right places.
I have added utility functions for working with precision in terms of
decimal significant figures as well as this code's native binary.
Cleanup the remnants of MS Compiler bits from <isc/refcount.h>, printing
the information in named/main.c, and cleanup some comments about Windows
that no longer apply.
The bits in picohttpparser.{h,c} were left out, because it's not our
code.
When fatal is called we may be holding memory allocated by OpenSSL.
This may result in the reference count for the FIPS provider not
going to zero and the shared library not being unloaded during
OPENSSL_cleanup. When the shared library is ultimately unloaded,
when all remaining dynamically loaded libraries are freed, we have
already destroyed the memory context we where using to track memory
leaks / late frees resulting in INSIST being called.
Disable triggering the INSIST when fatal has being called.
The `isc_trampoline` module had a lot of machinery to support stable
thread IDs for use by hazard pointers. But the hazard pointer code
is gone, and the `isc_loop` module now has its own per-loop thread
IDs.
The trampoline machinery seems over-complicated for its remaining
tasks, so move the per-thread initialization into `isc/thread.c`,
and delete the rest.
The isc_time_now() and isc_time_now_hires() were used inconsistently
through the code - either with status check, or without status check,
or via TIME_NOW() macro with RUNTIME_CHECK() on failure.
Refactor the isc_time_now() and isc_time_now_hires() to always fail when
getting current time has failed, and return the isc_time_t value as
return value instead of passing the pointer to result in the argument.
The isc_fsaccess API was created to hide the implementation details
between POSIX and Windows APIs. As we are not supporting the Windows
APIs anymore, it's better to drop this API used in the DST part.
Moreover, the isc_fsaccess was setting the permissions in an insecure
manner - it operated on the filename, and not on the file descriptor
which can lead to all kind of attacks if unpriviledged user has read (or
even worse write) access to key directory.
Replace the code that operates on the private keys with code that uses
mkstemp(), fchmod() and atomic rename() at the end, so at no time the
private key files have insecure permissions.
As it's impossible to get the current umask without modifying it at the
same time, initialize the current umask at the program start and keep
the loaded value internally. Add isc_os_umask() function to access the
starttime umask.
Instead of marking the unused entities with UNUSED(x) macro in the
function body, use a `ISC_ATTR_UNUSED` attribute macro that expans to
C23 [[maybe_unused]] or __attribute__((__unused__)) as fallback.
Use C23 attribute styles if available:
* Add new ISC_ATTR_UNUSED attribute macro that either expands to C23's
[[maybe_unused]] or __attribute__((__unused__));
* Add default expansion of the `noreturn` to [[noreturn]] if available;
* Move the FALLTHROUGH from <isc/util.h> to <isc/attributes.h>
Previously, isc_job_run() could have been used to run the job on the
current loop and the isc_job_run() would take care of allocating and
deallocating the job. After the change in this MR, the isc_job_run()
is more complicated to use, so we introduce the isc_async_current()
macro to suplement isc_async_run() when we need to run the job on the
current loop.
Change the isc_job_run() to not-make any allocations. The caller must
make sure that it allocates isc_job_t - usually as part of the argument
passed to the callback.
For simple jobs, using isc_async_run() is advised as it allocates its
own separate isc_job_t.
It's sometimes helpful to get a quick idea of the call stack when
debugging. This change factors out the backtrace logging from named's
fatal error handler so that it's easy to use in other places too.
when isc_nm_listenstreamdns() is called with a local port of 0,
a random port is chosen. call uv_getsockname() to determine what
the port is as soon as the socket is bound, and add a function
isc_nmsocket_getaddr() to retrieve it, so that the caller can
connect to the listening socket. this will be used in cases
where the same process is acting as both client and server.
ISC_REFCOUNT_TRACE_IMPL uses isc_tid(), but the corresponding header
file is not included, which breaks, for example, compiling BIND with
DNS_CATZ_TRACE defined in lib/dns/include/dns/catz.h.
Add '#include <isc/tid.h>' in lib/isc/include/isc/refcount.h.
This "quiescent state based reclamation" module provides support for
the qp-trie module in dns/qp. It is a replacement for liburcu, written
without reference to the urcu source code, and in fact it works in a
significantly different way.
A few specifics of BIND make this variant of QSBR somewhat simpler:
* We can require that wait-free access to a qp-trie only happens in
an isc_loop callback. The loop provides a natural quiescent state,
after the callbacks are done, when no qp-trie access occurs.
* We can dispense with any API like rcu_synchronize(). In practice,
it takes far too long to wait for a grace period to elapse for each
write to a data structure.
* We use the idea of "phases" (aka epochs or eras) from EBR to
reduce the amount of bookkeeping needed to track memory that is no
longer needed, knowing that the qp-trie does most of that work
already.
I considered hazard pointers for safe memory reclamation. They have
more read-side overhead (updating the hazard pointers) and it wasn't
clear to me how to nicely schedule the cleanup work. Another
alternative, epoch-based reclamation, is designed for fine-grained
lock-free updates, so it needs some rethinking to work well with the
heavily read-biased design of the qp-trie. QSBR has the fastest read
side of the basic SMR algorithms (with no barriers), and fits well
into a libuv loop. More recent hybrid SMR algorithms do not appear to
have enough benefits to justify the extra complexity.
the isc_glob module was originally needed to support posix-style glob
processing on Windows, but is now just an unnecessary wrapper around
glob(3). this commit removes it.
Add a singly-linked stack that supports lock-free prepend and drain (to
empty the list and clean up its elements). Intended for use with QSBR
to collect objects that need safe memory reclamation, or any other user
that works with adding objects to the stack and then draining them in
one go like various work queues.
In <isc/atomic.h>, add an `atomic_ptr()` macro to make type
declarations a little less abominable, and clean up a duplicate
definition of `atomic_compare_exchange_strong_acq_rel()`
removed references in code comments, doc/dev documentation, etc, to
isc_task, isc_timer_reset(), and isc_timertype_inactive. also removed a
coccinelle patch related to isc_timer_reset() that was no longer needed.
as there is no further use of isc_task in BIND, this commit removes
it, along with isc_taskmgr, isc_event, and all other related types.
functions that accepted taskmgr as a parameter have been cleaned up.
as a result of this change, some functions can no longer fail, so
they've been changed to type void, and their callers have been
updated accordingly.
the tasks table has been removed from the statistics channel and
the stats version has been updated. dns_dyndbctx has been changed
to reference the loopmgr instead of taskmgr, and DNS_DYNDB_VERSION
has been udpated as well.
change functions using isc_taskmgr_beginexclusive() to use
isc_loopmgr_pause() instead.
also, removed an unnecessary use of exclusive mode in
named_server_tcptimeouts().
most functions that were implemented as task events because they needed
to be running in a task to use exclusive mode have now been changed
into loop callbacks instead. (the exception is catz, which is being
changed in a separate commit because it's a particularly complex change.)
This changes the internal isc_rwlock implementation to:
Irina Calciu, Dave Dice, Yossi Lev, Victor Luchangco, Virendra
J. Marathe, and Nir Shavit. 2013. NUMA-aware reader-writer locks.
SIGPLAN Not. 48, 8 (August 2013), 157–166.
DOI:https://doi.org/10.1145/2517327.24425
(The full article available from:
http://mcg.cs.tau.ac.il/papers/ppopp2013-rwlocks.pdf)
The implementation is based on the The Writer-Preference Lock (C-RW-WP)
variant (see the 3.4 section of the paper for the rationale).
The implemented algorithm has been modified for simplicity and for usage
patterns in rbtdb.c.
The changes compared to the original algorithm:
* We haven't implemented the cohort locks because that would require a
knowledge of NUMA nodes, instead a simple atomic_bool is used as
synchronization point for writer lock.
* The per-thread reader counters are not being used - this would
require the internal thread id (isc_tid_v) to be always initialized,
even in the utilities; the change has a slight performance penalty,
so we might revisit this change in the future. However, this change
also saves a lot of memory, because cache-line aligned counters were
used, so on 32-core machine, the rwlock would be 4096+ bytes big.
* The readers use a writer_barrier that will raise after a while when
readers lock can't be acquired to prevent readers starvation.
* Separate ingress and egress readers counters queues to reduce both
inter and intra-thread contention.
Unfortunately, C still lacks a standard function for pause (x86,
sparc) or yeild (arm) instructions, for use in spin lock or CAS loops.
BIND has its own based on vendor intrinsics or inline asm.
Previously, it was buried in the `isc_rwlock` implementation. This
commit renames `isc_rwlock_pause()` to `isc_pause()` and moves
it into <isc/pause.h>.
This commit also fixes the configure script so that it detects ARM
yield support on systems that identify as `aarch*` instead of `arm*`.
On 64-bit ARM systems we now use the ISB (instruction synchronization
barrier) instruction in preference to yield. The ISB instruction
pauses the CPU for longer, several nanoseconds, which is more like the
x86 pause instruction. There are more details in a Rust pull request,
which also refers to MySQL making the same change:
https://github.com/rust-lang/rust/pull/84725
