Qpzone employs a locking strategy where rwlocks are grouped into
buckets, and each zone gets 17 buckets.
This strategy is suboptimal in two ways:
- If named is serving a single zone or a zone is the majority of the
traffic, this strategy pretty much guarantees contention when using
more than a dozen threads.
- If named is serving many small zones, it causes substantial memory
usage.
This commit switches the locking to a global table initialized at start
time. This should have three effects:
- Performance should improve in the single zone case, since now we are
selecting from a bigger pool of locks.
- Memory consumption should go down significantly in the many zone
cases.
- Performance should not degrade substantially in the many zone cases.
The reason for this is that, while we could have substantially more
zones than locks, we can query/edit only O(num threads) at the same
time. So by making the global table much bigger than the expected
number of threads, we can limit contention.
Instead of relying on unreliable order of execution of the library
constructors and destructors, move them to individual binaries. The
advantage is that the execution time and order will remain constant and
will not depend on the dynamic load dependency solver.
This requires more work, but that was mitigated by a simple requirement,
any executable using libisc and libdns, must include <isc/lib.h> and
<dns/lib.h> respectively (in this particular order). In turn, these two
headers must not be included from within any library as they contain
inlined functions marked with constructor/destructor attributes.
as libdns is no longer exported, it's not necessary to have
init and shutdown functions. the only purpose they served
was to create a private mctx and run dst_lib_init(), which
can be called directly instead.
The 'ephemeral' database implementation was used to provide a
lightweight database implemenation that doesn't cache results, and the
only place where it was really use is "samples" because delv is
overriding this to use "rbtdb" instead. Otherwise it was completely
unused.
* The 'ephemeral' cache DB (ecdb) implementation. An ecdb just provides
* temporary storage for ongoing name resolution with the common DB interfaces.
* It actually doesn't cache anything. The implementation expects any stored
* data is released within a short period, and does not care about the
* scalability in terms of the number of nodes.
The three functions has been modeled after the arc4random family of
functions, and they will always return random bytes.
The isc_random family of functions internally use these CSPRNG (if available):
1. getrandom() libc call (might be available on Linux and Solaris)
2. SYS_getrandom syscall (might be available on Linux, detected at runtime)
3. arc4random(), arc4random_buf() and arc4random_uniform() (available on BSDs and Mac OS X)
4. crypto library function:
4a. RAND_bytes in case OpenSSL
4b. pkcs_C_GenerateRandom() in case PKCS#11 library
4224. [func] Added support for "dyndb", a new interface for loading
zone data from an external database, developed by
Red Hat for the FreeIPA project.
DynDB drivers fully implement the BIND database
API, and are capable of significantly better
performance and functionality than DLZ drivers,
while taking advantage of advanced database
features not available in BIND such as multi-master
replication.
Thanks to Adam Tkac and Petr Spacek of Red Hat.
[RT #35271]
3760. [bug] Improve SIT with native PKCS#11 and on Windows.
[RT #35433]
3759. [port] Enable delve on Windows. [RT #35441]
3758. [port] Enable export library APIs on windows. [RT #35382]
iteration self tuning. The covers nodes clean from
the cache per iteration, nodes written to disk when
rewriting a master file and nodes destroyed per
iteration when destroying a zone or a cache.
[RT #14996]
Cleanup of redundant/useless header file inclusion.
ISC style lint, primarily for function declarations and standalone
comments -- ie, those that appear on a line without any code, which
should be written as follows:
/*
* This is a comment.
*/
