To allow testing for general support for fast bitscan intrinsics,
add symbols HAVE_BITSCAN_REVERSE and HAVE_BITSCAN_FORWARD.
Also do related cleanup in AllocSetFreeIndex(): Previously, we
tested for HAVE__BUILTIN_CLZ and copied the relevant internals of
pg_leftmost_one_pos32(), with a special fallback that does less
work than the general fallback for that function. Now that we have
a more general test, we just call pg_leftmost_one_pos32() directly
for platforms with intrinsic support. On gcc at least, there is no
difference in the binary for non-assert builds.
Discussion: https://www.postgresql.org/message-id/CAFBsxsEPc%2BFnX_0vmmQ5DHv60sk4rL_RZJ%2BMD6ei%3D76L0kFMvA%40mail.gmail.com
There were a number of places in the code that used bespoke bit-twiddling
expressions to do bitwise rotation. While we've had pg_rotate_right32()
for a while now, we hadn't gotten around to standardizing on that. Do so
now. Since many potential call sites look more natural with the "left"
equivalent, add that function too.
Reviewed by Tom Lane and Yugo Nagata
Discussion:
https://www.postgresql.org/message-id/CAFBsxsH7c1LC0CGZ0ADCBXLHU5-%3DKNXx-r7tHYPAW51b2HK4Qw%40mail.gmail.com
This reverts commits c2d1eea9e and 11b500072, as well as similar hacks
elsewhere, in favor of setting up the PGDLLIMPORT macro so that it can
just be used unconditionally. That can work because in frontend code,
we need no marking in either the defining or consuming files for a
variable exported from these libraries; and frontend code has no need
to access variables exported from the core backend, either.
While at it, write some actual documentation about the PGDLLIMPORT
and PGDLLEXPORT macros.
Patch by me, based on a suggestion from Robert Haas.
Discussion: https://postgr.es/m/1160385.1638165449@sss.pgh.pa.us
The point of introducing the hash_mem_multiplier GUC was to let users
reproduce the old behavior of hash aggregation, i.e. that it could use
more than work_mem at need. However, the implementation failed to get
the job done on Win64, where work_mem is clamped to 2GB to protect
various places that calculate memory sizes using "long int". As
written, the same clamp was applied to hash_mem. This resulted in
severe performance regressions for queries requiring a bit more than
2GB for hash aggregation, as they now spill to disk and there's no
way to stop that.
Getting rid of the work_mem restriction seems like a good idea, but
it's a big job and could not conceivably be back-patched. However,
there's only a fairly small number of places that are concerned with
the hash_mem value, and it turns out to be possible to remove the
restriction there without too much code churn or any ABI breaks.
So, let's do that for now to fix the regression, and leave the
larger task for another day.
This patch does introduce a bit more infrastructure that should help
with the larger task, namely pg_bitutils.h support for working with
size_t values.
Per gripe from Laurent Hasson. Back-patch to v13 where the
behavior change came in.
Discussion: https://postgr.es/m/997817.1627074924@sss.pgh.pa.us
Discussion: https://postgr.es/m/MN2PR15MB25601E80A9B6D1BA6F592B1985E39@MN2PR15MB2560.namprd15.prod.outlook.com
First pass of modifying various places that obtain the next power of 2 of
a number and make them use the new functions added in pg_bitutils.h
instead.
This also removes the _hash_log2() function. There are no longer any
callers in core. Other users can swap their _hash_log2(n) call to make use
of pg_ceil_log2_32(n).
Author: David Fetter, with some minor adjustments by me
Reviewed-by: John Naylor, Jesse Zhang
Discussion: https://postgr.es/m/20200114173553.GE32763%40fetter.org
There are many areas in the code where we need to determine the next
highest power of 2 of a given number. We tend to always do that in an
ad-hoc way each time, generally with some tight for loop which performs a
bitshift left once per loop and goes until it finds a number above the
given number.
Here we add two generic functions which make use of the existing
pg_leftmost_one_pos* functions which, when available, will allow us to
calculate the next power of 2 without any looping.
Here we don't add any code which uses these new functions. That will be
done in follow-up commits.
Author: David Fetter, with some minor adjustments by me
Reviewed-by: John Naylor, Jesse Zhang
Discussion: https://postgr.es/m/20200114173553.GE32763%40fetter.org
Our algorithm for choosing batch numbers turned out not to work
effectively for multi-billion key inner relations. We would use
more hash bits than we have, and effectively concentrate all tuples
into a smaller number of batches than we intended. While ideally
we should switch to wider hashes, for now, change the algorithm to
one that effectively gives up bits from the bucket number when we
don't have enough bits. That means we'll finish up with longer
bucket chains than would be ideal, but that's better than having
batches that don't fit in work_mem and can't be divided.
Batch-patch to all supported releases.
Author: Thomas Munro
Reviewed-by: Tom Lane, thanks also to Tomas Vondra, Alvaro Herrera, Andres Freund for testing and discussion
Reported-by: James Coleman
Discussion: https://postgr.es/m/16104-dc11ed911f1ab9df%40postgresql.org
Test for the compiler builtins __builtin_clz, __builtin_ctz, and
__builtin_popcount, and make use of these in preference to
handwritten C code if they're available. Create src/port
infrastructure for "leftmost one", "rightmost one", and "popcount"
so as to centralize these decisions.
On x86_64, __builtin_popcount generally won't make use of the POPCNT
opcode because that's not universally supported yet. Provide code
that checks CPUID and then calls POPCNT via asm() if available.
This requires indirecting through a function pointer, which is
an annoying amount of overhead for a one-instruction operation,
but it's probably not worth working harder than this for our
current use-cases.
I'm not sure we've found all the existing places that could profit
from this new infrastructure; but we at least touched all the
ones that used copied-and-pasted versions of the bitmapset.c code,
and got rid of multiple copies of the associated constant arrays.
While at it, replace c-compiler.m4's one-per-builtin-function
macros with a single one that can handle all the cases we need
to worry about so far. Also, because I'm paranoid, make those
checks into AC_LINK checks rather than just AC_COMPILE; the
former coding failed to verify that libgcc has support for the
builtin, in cases where it's not inline code.
David Rowley, Thomas Munro, Alvaro Herrera, Tom Lane
Discussion: https://postgr.es/m/CAKJS1f9WTAGG1tPeJnD18hiQW5gAk59fQ6WK-vfdAKEHyRg2RA@mail.gmail.com
This reverts commits fc6c72747a, 109de05cbb, d0b4663c23 and
711bab1e4d.
Somebody will have to try harder before submitting this patch again.
I've spent entirely too much time on it already, and the #ifdef maze yet
to be written in order for it to build at all got on my nerves. The
amount of work needed to get a platform-specific performance improvement
that's barely above the noise level is not worth it.
Split out these new functions in three parts: one in a new file that
uses the compiler builtin and gets compiled with the -mpopcnt compiler
option if it exists; another one that uses the compiler builtin but not
the compiler option; and finally the fallback with open-coded
algorithms.
Split out the configure logic: in the original commit, it was selecting
to use the -mpopcnt compiler switch together with deciding whether to
use the compiler builtin, but those two things are really separate.
Split them out. Also, expose whether the builtin exists to
Makefile.global, so that src/port's Makefile can decide whether to
compile the hw-optimized file.
Remove CPUID test for CTZ/CLZ. Make pg_{right,left}most_ones use either
the compiler intrinsic or open-coded algo; trying to use the
HW-optimized version is a waste of time. Make them static inline
functions.
Discussion: https://postgr.es/m/20190213221719.GA15976@alvherre.pgsql
These opcodes have been around in the AMD world since 2007, and 2008 in
the case of intel. They're supported in GCC and Clang via some __builtin
macros. The opcodes may be unavailable during runtime, in which case we
fall back on a C-based implementation of the code. In order to get the
POPCNT instruction we must pass the -mpopcnt option to the compiler. We
do this only for the pg_bitutils.c file.
David Rowley (with fragments taken from a patch by Thomas Munro)
Discussion: https://postgr.es/m/CAKJS1f9WTAGG1tPeJnD18hiQW5gAk59fQ6WK-vfdAKEHyRg2RA@mail.gmail.com
John Naylor
Bruce Momjian
Thomas Munro
Tom Lane
David Rowley
Michael Paquier
Alvaro Herrera