ARM platforms. This patch resolves this issue and unifies AUX column
processing at x86 and ARM using tempate class SimdProcessor.
The patch also replaces uint16_t mask previously used in column.cpp and
SimProcessor code with a native masks that platform uses, e.g. __m128i
or __m128 on x86 and variety of masks on ARM.
To unify the processing I introduced a new filtering Compare Operator - COMPARE_NULLEQ.
with a 'c1 IS NULL semantics'.
* more build dependencies
* fix for cmake < 3.11
It cannot do ADD_LIBRARY(... ALIAS ...) on IMPORTED targets
* another fix for cmake 3.10.2
It doesn't know about CMAKE_CXX_STANDARD=20,
let's add the correct flag manually
* gcc 8 on aarch64
utils/common/simd_arm.h:241:16: error: need ‘typename’ before ‘simd::TypeToVecWrapperType<T>::WrapperType’ because ‘simd::TypeToVecWrapperType<T>’ is a dependent scope
JobList low-level code relateod to primitive jobs now uses shared pointers instead of ByteStream refs talking to DEC
b/c same-node EM-PP communication now goes over a queue in DEC instead of a network hop.
PP now has a separate thread that processes the primitive job messages from that DEC queue.
Despite we have another number of tests in result, they all still run
gtests_add_test cannot parse TYPED_TEST_SUITE one by one and run them
in one bunch
operations + morsel size weight model to equally allocate CPU b/w parallel query morsels.
This patch delivers better parallel query timings distribution(timings graph resembles normal
distribution with a bigger left side thus more queries runs faster comparing with PrioThreadPool-based
single-node installation).
See changes in batchprimitiveprocessor-jl.h and comments in fair_threadpool.h for
important implementation details
PP now uses PriorityThreadPool that arbitrary picks another jobs pack
to run. This scheduling discipline tend to run portions of a single query
forcing other simultaneous queries to wait. In result parallel queries
timings variance is high. The FairThreadPool picks the job with the smallest
amount of work done so far(see the code for details)
EM scaleability project has two parts: phase1 and phase2.
This is phase1 that brings EM index to speed up(from O(n) down
to the speed of boost::unordered_map) EM lookups looking for
<dbroot, oid, partition> tuple to turn it into LBID,
e.g. most bulk insertion meta info operations.
The basis is boost::shared_managed_object where EMIndex is
stored. Whilst it is not debug-friendly it allows to put a
nested structs into shmem. EMIndex has 3 tiers. Top down description:
vector of dbroots, map of oids to partition vectors, partition
vectors that have EM indices.
Separate EM methods now queries index before they do EM run.
EMIndex has a separate shmem file with the fixed id
MCS-shm-00060001.
Short CHAR/VARCHAR column values contain integer-encoded strings.
After certain manipulations(orderSwap(strnxfrm(str))) the values
become integers that preserve original strings order relation
according to a certain translation rules(collation). Prepared
values are ready to be SIMD-processed.
* Fix clang warnings
* Remove vim tab guides
* initialize variables
* 'strncpy' output truncated before terminating nul copying as many bytes from a string as its length
* Fix ISO C++17 does not allow 'register' storage class specifier for outdated bison
* chars are unsigned on ARM, having if (ival < 0) always false
* chars are unsigned by default on ARM and comparison with -1 if always true
data types TEXT, CHAR, VARCHAR, FLOAT and DOUBLE are not yet supported by vectorized path
This patch introduces an example for Google benchmarking suite to measure a perf diff
b/w legacy scan/filtering code and the templated version
* Adds CompressInterfaceLZ4 which uses LZ4 API for compress/uncompress.
* Adds CMake machinery to search LZ4 on running host.
* All methods which use static data and do not modify any internal data - become `static`,
so we can use them without creation of the specific object. This is possible, because
the header specification has not been modified. We still use 2 sections in header, first
one with file meta data, the second one with pointers for compressed chunks.
* Methods `compress`, `uncompress`, `maxCompressedSize`, `getUncompressedSize` - become
pure virtual, so we can override them for the other compression algos.
* Adds method `getChunkMagicNumber`, so we can verify chunk magic number
for each compression algo.
* Renames "s/IDBCompressInterface/CompressInterface/g" according to requirement.
* This patch adds rebuildEM tool support to work with compressed files.
* This patch increases a version of the file header.
Note: Default version of the `rebuildEM` tool was using very old API,
those functions are not present currently. So `rebuildEM` will not work with
files created without compression, because we cannot deduce some info which are
needed to create column extent.
