Обсуждение: In PG12, query with float calculations is slower than PG11
Hi,
I am testing performance both PG12 and PG11.
I found the case of performance degradation in PG12.
Amit Langote help me to analyze and to create patch.
Thanks!
* environment
CentOS Linux release 7.6.1810 (Core)
postgresql 12.1
postgresql 11.6
* postgresql.conf
shared_buffers = 2048MB
max_parallel_workers_per_gather = 0
work_mem = '64MB'
jit = off
* test case
CREATE TABLE realtest(a real, b real, c real, d real, e real);
INSERT INTO realtest SELECT i,i,i,i,i FROM generate_series(0,10000000) AS i;
EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
* result
PG12.1 5878.389 ms
PG11.6 4533.554 ms
** PostgreSQL 12.1
pgbench=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------
Seq Scan on public.realtest (cost=0.00..288697.59 rows=10000115 width=40) (actual time=0.040..5195.328 rows=10000001 loops=1)
Output: ('2'::double precision * a), ('2'::double precision * b), ('2'::double precision * c), ('2'::double precision * d), ('2'::double precision * e)
Buffers: shared hit=63695
Planning Time: 0.051 ms
Execution Time: 5878.389 ms
(5 行)
Samples: 6K of event 'cpu-clock', Event count (approx.): 1577750000
Overhead Command Shared Object Symbol
25.48% postgres postgres [.] ExecInterpExpr
★18.65% postgres libc-2.17.so [.] __isinf
14.36% postgres postgres [.] float84mul
8.54% postgres [vdso] [.] __vdso_clock_gettime
4.02% postgres postgres [.] ExecScan
3.69% postgres postgres [.] tts_buffer_heap_getsomeattrs
2.63% postgres libc-2.17.so [.] __clock_gettime
2.55% postgres postgres [.] HeapTupleSatisfiesVisibility
2.00% postgres postgres [.] heapgettup_pagemode
** PostgreSQL 11.6
pgbench=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------
Seq Scan on public.realtest (cost=0.00..288697.59 rows=10000115 width=40) (actual time=0.012..3845.480 rows=10000001 loops=1)
Output: ('2'::double precision * a), ('2'::double precision * b), ('2'::double precision * c), ('2'::double precision * d), ('2'::double precision * e)
Buffers: shared hit=63695
Planning Time: 0.033 ms
Execution Time: 4533.554 ms
(5 行)
Samples: 4K of event 'cpu-clock', Event count (approx.): 1192000000
Overhead Command Shared Object Symbol
32.30% postgres postgres [.] ExecInterpExpr
14.95% postgres postgres [.] float84mul
10.57% postgres [vdso] [.] __vdso_clock_gettime
★ 6.84% postgres libc-2.17.so [.] __isinf
3.96% postgres postgres [.] ExecScan
3.50% postgres libc-2.17.so [.] __clock_gettime
3.31% postgres postgres [.] heap_getnext
3.08% postgres postgres [.] HeapTupleSatisfiesMVCC
2.77% postgres postgres [.] slot_deform_tuple
2.37% postgres postgres [.] ExecProcNodeInstr
2.08% postgres postgres [.] standard_ExecutorRun
* cause
Obviously, even in common cases where no overflow occurs,
you can tell that PG 12 is performing isinf() 3 times on every call of
float8_mul() once for each of val1, val2, result where as PG 11
is performing only once for result.
That's because check_float8_val() (in PG 12) is a function
whose arguments must be evaluated before
it is called (it is inline, but that's irrelevant),
whereas CHECKFLOATVAL() (in PG11) is a macro
whose arguments are only substituted into its body.
By the way, this change of float8mul() implementation is
mostly due to the following commit in PG 12 development cycle:
commit 6bf0bc842bd75877e31727eb559c6a69e237f831
Especially, the following diff:
@@ -894,13 +746,8 @@ float8mul(PG_FUNCTION_ARGS) {
float8 arg1 = PG_GETARG_FLOAT8(0);
float8 arg2 = PG_GETARG_FLOAT8(1);
- float8 result;
-
- result = arg1 * arg2;
- CHECKFLOATVAL(result, isinf(arg1) || isinf(arg2),
- arg1 == 0 || arg2 == 0);
- PG_RETURN_FLOAT8(result);
+ PG_RETURN_FLOAT8(float8_mul(arg1, arg2));
}
* patch
This patch uses MACRO which was used by PG11.
I tried attached patch, which can be applied to PG 12 source and performed a benchmark:
PG12.1 5878.389 ms
PG11.6 4533.554 ms
PG12.1 + Patch 4679.162 ms
** PostgreSQL 12.1 + Patch
postgres=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------
Seq Scan on public.realtest (cost=0.00..307328.38 rows=10828150 width=40) (actual time=0.012..4009.012 rows=10000001 loops=1)
Output: ('2'::double precision * a), ('2'::double precision * b), ('2'::double precision * c), ('2'::double precision * d), ('2'::double precision * e)
Buffers: shared hit=63695
Planning Time: 0.038 ms
Execution Time: 4679.162 ms
(5 rows)
Samples: 5K of event 'cpu-clock', Event count (approx.): 1376750000
Overhead Command Shared Object Symbol
31.43% postgres postgres [.] ExecInterpExpr
14.24% postgres postgres [.] float84mul
10.40% postgres [vdso] [.] __vdso_clock_gettime
★ 5.41% postgres libc-2.17.so [.] __isinf
4.63% postgres postgres [.] tts_buffer_heap_getsomeattrs
4.03% postgres postgres [.] ExecScan
3.54% postgres libc-2.17.so [.] __clock_gettime
3.12% postgres postgres [.] HeapTupleSatisfiesVisibility
2.36% postgres postgres [.] heap_getnextslot
2.16% postgres postgres [.] heapgettup_pagemode
2.09% postgres postgres [.] standard_ExecutorRun
2.07% postgres postgres [.] SeqNext
2.03% postgres postgres [.] ExecProcNodeInstr
2.03% postgres postgres [.] tts_virtual_clear
PG 12 is still slower compared to PG 11, but the __isinf() situation is better with the patch.
Best Regards,
Keisuke Kuroda
I am testing performance both PG12 and PG11.
I found the case of performance degradation in PG12.
Amit Langote help me to analyze and to create patch.
Thanks!
* environment
CentOS Linux release 7.6.1810 (Core)
postgresql 12.1
postgresql 11.6
* postgresql.conf
shared_buffers = 2048MB
max_parallel_workers_per_gather = 0
work_mem = '64MB'
jit = off
* test case
CREATE TABLE realtest(a real, b real, c real, d real, e real);
INSERT INTO realtest SELECT i,i,i,i,i FROM generate_series(0,10000000) AS i;
EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
* result
PG12.1 5878.389 ms
PG11.6 4533.554 ms
** PostgreSQL 12.1
pgbench=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------
Seq Scan on public.realtest (cost=0.00..288697.59 rows=10000115 width=40) (actual time=0.040..5195.328 rows=10000001 loops=1)
Output: ('2'::double precision * a), ('2'::double precision * b), ('2'::double precision * c), ('2'::double precision * d), ('2'::double precision * e)
Buffers: shared hit=63695
Planning Time: 0.051 ms
Execution Time: 5878.389 ms
(5 行)
Samples: 6K of event 'cpu-clock', Event count (approx.): 1577750000
Overhead Command Shared Object Symbol
25.48% postgres postgres [.] ExecInterpExpr
★18.65% postgres libc-2.17.so [.] __isinf
14.36% postgres postgres [.] float84mul
8.54% postgres [vdso] [.] __vdso_clock_gettime
4.02% postgres postgres [.] ExecScan
3.69% postgres postgres [.] tts_buffer_heap_getsomeattrs
2.63% postgres libc-2.17.so [.] __clock_gettime
2.55% postgres postgres [.] HeapTupleSatisfiesVisibility
2.00% postgres postgres [.] heapgettup_pagemode
** PostgreSQL 11.6
pgbench=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------
Seq Scan on public.realtest (cost=0.00..288697.59 rows=10000115 width=40) (actual time=0.012..3845.480 rows=10000001 loops=1)
Output: ('2'::double precision * a), ('2'::double precision * b), ('2'::double precision * c), ('2'::double precision * d), ('2'::double precision * e)
Buffers: shared hit=63695
Planning Time: 0.033 ms
Execution Time: 4533.554 ms
(5 行)
Samples: 4K of event 'cpu-clock', Event count (approx.): 1192000000
Overhead Command Shared Object Symbol
32.30% postgres postgres [.] ExecInterpExpr
14.95% postgres postgres [.] float84mul
10.57% postgres [vdso] [.] __vdso_clock_gettime
★ 6.84% postgres libc-2.17.so [.] __isinf
3.96% postgres postgres [.] ExecScan
3.50% postgres libc-2.17.so [.] __clock_gettime
3.31% postgres postgres [.] heap_getnext
3.08% postgres postgres [.] HeapTupleSatisfiesMVCC
2.77% postgres postgres [.] slot_deform_tuple
2.37% postgres postgres [.] ExecProcNodeInstr
2.08% postgres postgres [.] standard_ExecutorRun
* cause
Obviously, even in common cases where no overflow occurs,
you can tell that PG 12 is performing isinf() 3 times on every call of
float8_mul() once for each of val1, val2, result where as PG 11
is performing only once for result.
That's because check_float8_val() (in PG 12) is a function
whose arguments must be evaluated before
it is called (it is inline, but that's irrelevant),
whereas CHECKFLOATVAL() (in PG11) is a macro
whose arguments are only substituted into its body.
By the way, this change of float8mul() implementation is
mostly due to the following commit in PG 12 development cycle:
commit 6bf0bc842bd75877e31727eb559c6a69e237f831
Especially, the following diff:
@@ -894,13 +746,8 @@ float8mul(PG_FUNCTION_ARGS) {
float8 arg1 = PG_GETARG_FLOAT8(0);
float8 arg2 = PG_GETARG_FLOAT8(1);
- float8 result;
-
- result = arg1 * arg2;
- CHECKFLOATVAL(result, isinf(arg1) || isinf(arg2),
- arg1 == 0 || arg2 == 0);
- PG_RETURN_FLOAT8(result);
+ PG_RETURN_FLOAT8(float8_mul(arg1, arg2));
}
* patch
This patch uses MACRO which was used by PG11.
I tried attached patch, which can be applied to PG 12 source and performed a benchmark:
PG12.1 5878.389 ms
PG11.6 4533.554 ms
PG12.1 + Patch 4679.162 ms
** PostgreSQL 12.1 + Patch
postgres=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------
Seq Scan on public.realtest (cost=0.00..307328.38 rows=10828150 width=40) (actual time=0.012..4009.012 rows=10000001 loops=1)
Output: ('2'::double precision * a), ('2'::double precision * b), ('2'::double precision * c), ('2'::double precision * d), ('2'::double precision * e)
Buffers: shared hit=63695
Planning Time: 0.038 ms
Execution Time: 4679.162 ms
(5 rows)
Samples: 5K of event 'cpu-clock', Event count (approx.): 1376750000
Overhead Command Shared Object Symbol
31.43% postgres postgres [.] ExecInterpExpr
14.24% postgres postgres [.] float84mul
10.40% postgres [vdso] [.] __vdso_clock_gettime
★ 5.41% postgres libc-2.17.so [.] __isinf
4.63% postgres postgres [.] tts_buffer_heap_getsomeattrs
4.03% postgres postgres [.] ExecScan
3.54% postgres libc-2.17.so [.] __clock_gettime
3.12% postgres postgres [.] HeapTupleSatisfiesVisibility
2.36% postgres postgres [.] heap_getnextslot
2.16% postgres postgres [.] heapgettup_pagemode
2.09% postgres postgres [.] standard_ExecutorRun
2.07% postgres postgres [.] SeqNext
2.03% postgres postgres [.] ExecProcNodeInstr
2.03% postgres postgres [.] tts_virtual_clear
PG 12 is still slower compared to PG 11, but the __isinf() situation is better with the patch.
Best Regards,
Keisuke Kuroda
Вложения
Hi, On 2020-02-06 14:25:03 +0900, keisuke kuroda wrote: > That's because check_float8_val() (in PG 12) is a function > whose arguments must be evaluated before > it is called (it is inline, but that's irrelevant), > whereas CHECKFLOATVAL() (in PG11) is a macro > whose arguments are only substituted into its body. Hm - it's not that clear to me that it is irrelevant that the function gets inlined. The compiler should know that isinf is side-effect free, and that it doesn't have to evaluate before necessary. Normally isinf is implemented by a compiler intrisic within the system headers. But not in your profile: > ★ 5.41% postgres libc-2.17.so [.] __isinf I checked, and I don't see any references to isinf from within float.c (looking at the disassembly - there's some debug strings containing the word, but that's it). What compiler & compiler version on what kind of architecture is this? Greetings, Andres Freund
Hi,
On Thu, Feb 6, 2020 at 2:55 PM Andres Freund <andres@anarazel.de> wrote:
> On 2020-02-06 14:25:03 +0900, keisuke kuroda wrote:
> > That's because check_float8_val() (in PG 12) is a function
> > whose arguments must be evaluated before
> > it is called (it is inline, but that's irrelevant),
> > whereas CHECKFLOATVAL() (in PG11) is a macro
> > whose arguments are only substituted into its body.
>
> Hm - it's not that clear to me that it is irrelevant that the function
> gets inlined. The compiler should know that isinf is side-effect free,
> and that it doesn't have to evaluate before necessary.
>
> Normally isinf is implemented by a compiler intrisic within the system
> headers. But not in your profile:
> > ★ 5.41% postgres libc-2.17.so [.] __isinf
>
> I checked, and I don't see any references to isinf from within float.c
> (looking at the disassembly - there's some debug strings containing the
> word, but that's it).
>
> What compiler & compiler version on what kind of architecture is this?
As Kuroda-san mentioned, I also checked the behavior that he reports.
The compiler I used is an ancient one (CentOS 7 default):
$ gcc --version
gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-39)
Compiler dependent behavior of inlining might be relevant here, but
there is one more thing to consider. The if () condition in
check_float8_val (PG 12) and CHECKFLOATVAL (PG 11) is calculated
differently, causing isinf() to be called more times in PG 12:
static inline void
check_float8_val(const float8 val, const bool inf_is_valid,
const bool zero_is_valid)
{
if (!inf_is_valid && unlikely(isinf(val)))
ereport(ERROR,
(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
errmsg("value out of range: overflow")));
#define CHECKFLOATVAL(val, inf_is_valid, zero_is_valid) \
do { \
if (isinf(val) && !(inf_is_valid)) \
ereport(ERROR, \
(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE), \
errmsg("value out of range: overflow"))); \
called thusly:
check_float8_val(result, isinf(val1) || isinf(val2),
val1 == 0.0 || val2 == 0.0);
and
CHECKFLOATVAL(result, isinf(arg1) || isinf(arg2),
arg1 == 0 || arg2 == 0);
from float8_mul() and float8mul() in PG 12 and PG 11, respectively.
You may notice that the if () condition is reversed, so while PG 12
calculates isinf(arg1) || isinf(arg2) first and isinf(result) only if
the first is false, which it is in most cases, PG 11 calculates
isinf(result) first, followed by isinf(arg1) || isinf(arg2) if the
former is true. I don't understand why such reversal was necessary,
but it appears to be the main factor behind this slowdown. So, even
if PG 12's check_float8_val() is perfectly inlined, this slowdown
couldn't be helped.
Thanks,
Amit
So it appears to me that what commit 6bf0bc842 did in this area was
not just wrong, but disastrously so. Before that, we had a macro that
evaluated isinf(val) before it evaluated the inf_is_valid condition.
Now we have check_float[48]_val which do it the other way around.
That would be okay if the inf_is_valid condition were cheap to
evaluate, but in common code paths it's actually twice as expensive
as isinf().
Andres seems to be of the opinion that the compiler should be willing
to ignore the semantic requirements of the C standard in order
to rearrange the code back into the cheaper order. That sounds like
wishful thinking to me ... even if it actually works on his compiler,
it certainly isn't going to work for everyone.
The patch looks unduly invasive to me, but I think that it might be
right that we should go back to a macro-based implementation, because
otherwise we don't have a good way to be certain that the function
parameter won't get evaluated first. (Another reason to do so is
so that the file/line numbers generated for the error reports go back
to being at least a little bit useful.) We could use local variables
within the macro to avoid double evals, if anyone thinks that's
actually important --- I don't.
I think the current code is probably also misusing unlikely(),
and that the right way would be more like
if (unlikely(isinf(val) && !(inf_is_valid)))
regards, tom lane
On Thu, Feb 6, 2020 at 11:04 AM Tom Lane <tgl@sss.pgh.pa.us> wrote: > So it appears to me that what commit 6bf0bc842 did in this area was > not just wrong, but disastrously so. Before that, we had a macro that > evaluated isinf(val) before it evaluated the inf_is_valid condition. > Now we have check_float[48]_val which do it the other way around. > That would be okay if the inf_is_valid condition were cheap to > evaluate, but in common code paths it's actually twice as expensive > as isinf(). Well, if the previous coding was a deliberate attempt to dodge this performance issue, the evidence seems to be well-concealed. Neither the comments for that macro nor the related commit messages make any mention of it. When subtle things like this are performance-critical, good comments are pretty critical, too. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company
Hi, On 2020-02-06 11:03:51 -0500, Tom Lane wrote: > Andres seems to be of the opinion that the compiler should be willing > to ignore the semantic requirements of the C standard in order > to rearrange the code back into the cheaper order. That sounds like > wishful thinking to me ... even if it actually works on his compiler, > it certainly isn't going to work for everyone. Sorry, but, uh, what are you talking about? Please tell me which single standards violation I'm advocating for? I was asking about the inlining bit because the first email of the topic explained that as the problem, which I don't believe can be the full explanation - and it turns out it isn't. As Amit Langote's followup email explained, there's the whole issue of the order of checks being inverted - which is clearly bad. And wholly unrelated to inlining. And I asked about __isinf() being used because there are issues with accidentally ending up with the non-intrinsic version of isinf() when not using gcc, due to badly written standard library headers. > The patch looks unduly invasive to me, but I think that it might be > right that we should go back to a macro-based implementation, because > otherwise we don't have a good way to be certain that the function > parameter won't get evaluated first. I'd first like to see some actual evidence of this being a problem, rather than just the order of the checks. > (Another reason to do so is so that the file/line numbers generated > for the error reports go back to being at least a little bit useful.) > We could use local variables within the macro to avoid double evals, > if anyone thinks that's actually important --- I don't. I don't think that's necessarily a good idea. In fact, I think we should probably do the exact opposite, and move the error messages further out of line. All these otherwise very small functions having their own ereports makes them much bigger. Our low code density, and the resulting rate of itlb misses, is pretty significant cost (cf [1]). master: text data bss dec hex filename 36124 44 65 36233 8d89 float.o error messages moved out of line: text data bss dec hex filename 32883 44 65 32992 80e0 float.o Taking int4pl as an example - solely because it is simpler assembly to look at - we get: master: 0x00000000004ac190 <+0>: mov 0x30(%rdi),%rax 0x00000000004ac194 <+4>: add 0x20(%rdi),%eax 0x00000000004ac197 <+7>: jo 0x4ac19c <int4pl+12> 0x00000000004ac199 <+9>: cltq 0x00000000004ac19b <+11>: retq 0x00000000004ac19c <+12>: push %rbp 0x00000000004ac19d <+13>: lea 0x1a02c4(%rip),%rsi # 0x64c468 0x00000000004ac1a4 <+20>: xor %r8d,%r8d 0x00000000004ac1a7 <+23>: lea 0x265da1(%rip),%rcx # 0x711f4f <__func__.26823> 0x00000000004ac1ae <+30>: mov $0x30b,%edx 0x00000000004ac1b3 <+35>: mov $0x14,%edi 0x00000000004ac1b8 <+40>: callq 0x586060 <errstart> 0x00000000004ac1bd <+45>: lea 0x147e0e(%rip),%rdi # 0x5f3fd2 0x00000000004ac1c4 <+52>: xor %eax,%eax 0x00000000004ac1c6 <+54>: callq 0x5896a0 <errmsg> 0x00000000004ac1cb <+59>: mov $0x3000082,%edi 0x00000000004ac1d0 <+64>: mov %eax,%ebp 0x00000000004ac1d2 <+66>: callq 0x589540 <errcode> 0x00000000004ac1d7 <+71>: mov %eax,%edi 0x00000000004ac1d9 <+73>: mov %ebp,%esi 0x00000000004ac1db <+75>: xor %eax,%eax 0x00000000004ac1dd <+77>: callq 0x588fb0 <errfinish> out-of-line error: 0x00000000004b04e0 <+0>: mov 0x30(%rdi),%rax 0x00000000004b04e4 <+4>: add 0x20(%rdi),%eax 0x00000000004b04e7 <+7>: jo 0x4b04ec <int4pl+12> 0x00000000004b04e9 <+9>: cltq 0x00000000004b04eb <+11>: retq 0x00000000004b04ec <+12>: push %rax 0x00000000004b04ed <+13>: callq 0x115e17 <out_of_range_err> With the out-of-line error, we can fit multiple of these functions into one cache line. With the inline error, not even one. Greetings, Andres Freund [1] https://twitter.com/AndresFreundTec/status/1214305610172289024
Hi,
I have been testing with newer compiler (clang-7)
and result is a bit different at least with clang-7.
Compiling PG 12.1 (even without patch) with clang-7
results in __isinf() no longer being a bottleneck,
that is, you don't see it in profiler at all.
So, there is no issue for people who use the modern clang toolchain,
but maybe that's not everyone.
So there would still be some interest in doing something about this.
* clang
bash-4.2$ which clang
/opt/rh/llvm-toolset-7.0/root/usr/bin/clang
bash-4.2$ clang -v
clang version 7.0.1 (tags/RELEASE_701/final)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/rh/llvm-toolset-7.0/root/usr/bin
Found candidate GCC installation: /opt/rh/devtoolset-7/root/usr/lib/gcc/x86_64-redhat-linux/7
Found candidate GCC installation: /opt/rh/devtoolset-8/root/usr/lib/gcc/x86_64-redhat-linux/8
Found candidate GCC installation: /usr/lib/gcc/x86_64-redhat-linux/4.8.2
Found candidate GCC installation: /usr/lib/gcc/x86_64-redhat-linux/4.8.5
Selected GCC installation: /opt/rh/devtoolset-8/root/usr/lib/gcc/x86_64-redhat-linux/8
Candidate multilib: .;@m64
Candidate multilib: 32;@m32
Selected multilib: .;@m64
** pg_config
---
CONFIGURE = '--prefix=/var/lib/pgsql/pgsql/12.1' 'CC=/opt/rh/llvm-toolset-7.0/root/usr/bin/clang' 'PKG_CONFIG_PATH=/opt/rh/llvm-toolset-7.0/root/usr/lib64/pkgconfig'
CC = /opt/rh/llvm-toolset-7.0/root/usr/bin/clang
---
* result(PostgreSQL 12.1 (even without patch))
postgres=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------
Seq Scan on public.realtest (cost=0.00..288697.59 rows=10000115 width=40) (actual time=0.012..3878.284 rows=10000001 loops=1)
Output: ('2'::double precision * a), ('2'::double precision * b), ('2'::double precision * c), ('2'::double precision * d), ('2'::double precision * e)
Buffers: shared hit=63695
Planning Time: 0.038 ms
Execution Time: 4533.767 ms
(5 rows)
Samples: 5K of event 'cpu-clock', Event count (approx.): 1275000000
Overhead Command Shared Object Symbol
33.92% postgres postgres [.] ExecInterpExpr
13.27% postgres postgres [.] float84mul
10.86% postgres [vdso] [.] __vdso_clock_gettime
5.49% postgres postgres [.] tts_buffer_heap_getsomeattrs
3.96% postgres postgres [.] ExecScan
3.25% postgres libc-2.17.so [.] __clock_gettime
3.16% postgres postgres [.] heap_getnextslot
2.41% postgres postgres [.] tts_virtual_clear
2.39% postgres postgres [.] SeqNext
2.22% postgres postgres [.] InstrStopNode
Best Regards,
Keisuke Kuroda
I have been testing with newer compiler (clang-7)
and result is a bit different at least with clang-7.
Compiling PG 12.1 (even without patch) with clang-7
results in __isinf() no longer being a bottleneck,
that is, you don't see it in profiler at all.
So, there is no issue for people who use the modern clang toolchain,
but maybe that's not everyone.
So there would still be some interest in doing something about this.
* clang
bash-4.2$ which clang
/opt/rh/llvm-toolset-7.0/root/usr/bin/clang
bash-4.2$ clang -v
clang version 7.0.1 (tags/RELEASE_701/final)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/rh/llvm-toolset-7.0/root/usr/bin
Found candidate GCC installation: /opt/rh/devtoolset-7/root/usr/lib/gcc/x86_64-redhat-linux/7
Found candidate GCC installation: /opt/rh/devtoolset-8/root/usr/lib/gcc/x86_64-redhat-linux/8
Found candidate GCC installation: /usr/lib/gcc/x86_64-redhat-linux/4.8.2
Found candidate GCC installation: /usr/lib/gcc/x86_64-redhat-linux/4.8.5
Selected GCC installation: /opt/rh/devtoolset-8/root/usr/lib/gcc/x86_64-redhat-linux/8
Candidate multilib: .;@m64
Candidate multilib: 32;@m32
Selected multilib: .;@m64
** pg_config
---
CONFIGURE = '--prefix=/var/lib/pgsql/pgsql/12.1' 'CC=/opt/rh/llvm-toolset-7.0/root/usr/bin/clang' 'PKG_CONFIG_PATH=/opt/rh/llvm-toolset-7.0/root/usr/lib64/pkgconfig'
CC = /opt/rh/llvm-toolset-7.0/root/usr/bin/clang
---
* result(PostgreSQL 12.1 (even without patch))
postgres=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------
Seq Scan on public.realtest (cost=0.00..288697.59 rows=10000115 width=40) (actual time=0.012..3878.284 rows=10000001 loops=1)
Output: ('2'::double precision * a), ('2'::double precision * b), ('2'::double precision * c), ('2'::double precision * d), ('2'::double precision * e)
Buffers: shared hit=63695
Planning Time: 0.038 ms
Execution Time: 4533.767 ms
(5 rows)
Samples: 5K of event 'cpu-clock', Event count (approx.): 1275000000
Overhead Command Shared Object Symbol
33.92% postgres postgres [.] ExecInterpExpr
13.27% postgres postgres [.] float84mul
10.86% postgres [vdso] [.] __vdso_clock_gettime
5.49% postgres postgres [.] tts_buffer_heap_getsomeattrs
3.96% postgres postgres [.] ExecScan
3.25% postgres libc-2.17.so [.] __clock_gettime
3.16% postgres postgres [.] heap_getnextslot
2.41% postgres postgres [.] tts_virtual_clear
2.39% postgres postgres [.] SeqNext
2.22% postgres postgres [.] InstrStopNode
Best Regards,
Keisuke Kuroda
2020年2月7日(金) 3:48 Andres Freund <andres@anarazel.de>:
Hi,
On 2020-02-06 11:03:51 -0500, Tom Lane wrote:
> Andres seems to be of the opinion that the compiler should be willing
> to ignore the semantic requirements of the C standard in order
> to rearrange the code back into the cheaper order. That sounds like
> wishful thinking to me ... even if it actually works on his compiler,
> it certainly isn't going to work for everyone.
Sorry, but, uh, what are you talking about? Please tell me which single
standards violation I'm advocating for?
I was asking about the inlining bit because the first email of the topic
explained that as the problem, which I don't believe can be the full
explanation - and it turns out it isn't. As Amit Langote's followup
email explained, there's the whole issue of the order of checks being
inverted - which is clearly bad. And wholly unrelated to inlining.
And I asked about __isinf() being used because there are issues with
accidentally ending up with the non-intrinsic version of isinf() when
not using gcc, due to badly written standard library headers.
> The patch looks unduly invasive to me, but I think that it might be
> right that we should go back to a macro-based implementation, because
> otherwise we don't have a good way to be certain that the function
> parameter won't get evaluated first.
I'd first like to see some actual evidence of this being a problem,
rather than just the order of the checks.
> (Another reason to do so is so that the file/line numbers generated
> for the error reports go back to being at least a little bit useful.)
> We could use local variables within the macro to avoid double evals,
> if anyone thinks that's actually important --- I don't.
I don't think that's necessarily a good idea. In fact, I think we should
probably do the exact opposite, and move the error messages further out
of line. All these otherwise very small functions having their own
ereports makes them much bigger. Our low code density, and the resulting
rate of itlb misses, is pretty significant cost (cf [1]).
master:
text data bss dec hex filename
36124 44 65 36233 8d89 float.o
error messages moved out of line:
text data bss dec hex filename
32883 44 65 32992 80e0 float.o
Taking int4pl as an example - solely because it is simpler assembly to
look at - we get:
master:
0x00000000004ac190 <+0>: mov 0x30(%rdi),%rax
0x00000000004ac194 <+4>: add 0x20(%rdi),%eax
0x00000000004ac197 <+7>: jo 0x4ac19c <int4pl+12>
0x00000000004ac199 <+9>: cltq
0x00000000004ac19b <+11>: retq
0x00000000004ac19c <+12>: push %rbp
0x00000000004ac19d <+13>: lea 0x1a02c4(%rip),%rsi # 0x64c468
0x00000000004ac1a4 <+20>: xor %r8d,%r8d
0x00000000004ac1a7 <+23>: lea 0x265da1(%rip),%rcx # 0x711f4f <__func__.26823>
0x00000000004ac1ae <+30>: mov $0x30b,%edx
0x00000000004ac1b3 <+35>: mov $0x14,%edi
0x00000000004ac1b8 <+40>: callq 0x586060 <errstart>
0x00000000004ac1bd <+45>: lea 0x147e0e(%rip),%rdi # 0x5f3fd2
0x00000000004ac1c4 <+52>: xor %eax,%eax
0x00000000004ac1c6 <+54>: callq 0x5896a0 <errmsg>
0x00000000004ac1cb <+59>: mov $0x3000082,%edi
0x00000000004ac1d0 <+64>: mov %eax,%ebp
0x00000000004ac1d2 <+66>: callq 0x589540 <errcode>
0x00000000004ac1d7 <+71>: mov %eax,%edi
0x00000000004ac1d9 <+73>: mov %ebp,%esi
0x00000000004ac1db <+75>: xor %eax,%eax
0x00000000004ac1dd <+77>: callq 0x588fb0 <errfinish>
out-of-line error:
0x00000000004b04e0 <+0>: mov 0x30(%rdi),%rax
0x00000000004b04e4 <+4>: add 0x20(%rdi),%eax
0x00000000004b04e7 <+7>: jo 0x4b04ec <int4pl+12>
0x00000000004b04e9 <+9>: cltq
0x00000000004b04eb <+11>: retq
0x00000000004b04ec <+12>: push %rax
0x00000000004b04ed <+13>: callq 0x115e17 <out_of_range_err>
With the out-of-line error, we can fit multiple of these functions into one
cache line. With the inline error, not even one.
Greetings,
Andres Freund
[1] https://twitter.com/AndresFreundTec/status/1214305610172289024
Hi,
On February 6, 2020 11:42:30 PM PST, keisuke kuroda <keisuke.kuroda.3862@gmail.com> wrote:
>Hi,
>
>I have been testing with newer compiler (clang-7)
>and result is a bit different at least with clang-7.
>Compiling PG 12.1 (even without patch) with clang-7
>results in __isinf() no longer being a bottleneck,
>that is, you don't see it in profiler at all.
I don't think that's necessarily the right conclusion. What's quite possibly happening is that you do not see the
externalisinf function anymore, because it is implemented as an intrinsic, but that there still are more computations
beingdone. Due to the changed order of the isinf checks. You'd have to compare with 11 using the same compiler.
Andres
>* result(PostgreSQL 12.1 (even without patch))
>
>postgres=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
> select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
> from realtest;
>
>QUERY PLAN
>-----------------------------------------------------------------------------------------------------------------------
>Seq Scan on public.realtest (cost=0.00..288697.59 rows=10000115
>width=40)
>(actual time=0.012..3878.284 rows=10000001 loops=1)
> Output: ('2'::double precision * a), ('2'::double precision * b),
>('2'::double precision * c), ('2'::double precision * d), ('2'::double
>precision * e)
> Buffers: shared hit=63695
> Planning Time: 0.038 ms
> Execution Time: 4533.767 ms
>(5 rows)
>
>Samples: 5K of event 'cpu-clock', Event count (approx.): 1275000000
>Overhead Command Shared Object Symbol
> 33.92% postgres postgres [.] ExecInterpExpr
> 13.27% postgres postgres [.] float84mul
> 10.86% postgres [vdso] [.] __vdso_clock_gettime
> 5.49% postgres postgres [.] tts_buffer_heap_getsomeattrs
> 3.96% postgres postgres [.] ExecScan
> 3.25% postgres libc-2.17.so [.] __clock_gettime
> 3.16% postgres postgres [.] heap_getnextslot
> 2.41% postgres postgres [.] tts_virtual_clear
> 2.39% postgres postgres [.] SeqNext
> 2.22% postgres postgres [.] InstrStopNode
>
>Best Regards,
>Keisuke Kuroda
>
>2020年2月7日(金) 3:48 Andres Freund <andres@anarazel.de>:
>
>> Hi,
>>
>> On 2020-02-06 11:03:51 -0500, Tom Lane wrote:
>> > Andres seems to be of the opinion that the compiler should be
>willing
>> > to ignore the semantic requirements of the C standard in order
>> > to rearrange the code back into the cheaper order. That sounds
>like
>> > wishful thinking to me ... even if it actually works on his
>compiler,
>> > it certainly isn't going to work for everyone.
>>
>> Sorry, but, uh, what are you talking about? Please tell me which
>single
>> standards violation I'm advocating for?
>>
>> I was asking about the inlining bit because the first email of the
>topic
>> explained that as the problem, which I don't believe can be the full
>> explanation - and it turns out it isn't. As Amit Langote's followup
>> email explained, there's the whole issue of the order of checks being
>> inverted - which is clearly bad. And wholly unrelated to inlining.
>>
>> And I asked about __isinf() being used because there are issues with
>> accidentally ending up with the non-intrinsic version of isinf() when
>> not using gcc, due to badly written standard library headers.
>>
>>
>> > The patch looks unduly invasive to me, but I think that it might be
>> > right that we should go back to a macro-based implementation,
>because
>> > otherwise we don't have a good way to be certain that the function
>> > parameter won't get evaluated first.
>>
>> I'd first like to see some actual evidence of this being a problem,
>> rather than just the order of the checks.
>>
>>
>> > (Another reason to do so is so that the file/line numbers generated
>> > for the error reports go back to being at least a little bit
>useful.)
>> > We could use local variables within the macro to avoid double
>evals,
>> > if anyone thinks that's actually important --- I don't.
>>
>> I don't think that's necessarily a good idea. In fact, I think we
>should
>> probably do the exact opposite, and move the error messages further
>out
>> of line. All these otherwise very small functions having their own
>> ereports makes them much bigger. Our low code density, and the
>resulting
>> rate of itlb misses, is pretty significant cost (cf [1]).
>>
>> master:
>> text data bss dec hex filename
>> 36124 44 65 36233 8d89 float.o
>> error messages moved out of line:
>> text data bss dec hex filename
>> 32883 44 65 32992 80e0 float.o
>>
>> Taking int4pl as an example - solely because it is simpler assembly
>to
>> look at - we get:
>>
>> master:
>> 0x00000000004ac190 <+0>: mov 0x30(%rdi),%rax
>> 0x00000000004ac194 <+4>: add 0x20(%rdi),%eax
>> 0x00000000004ac197 <+7>: jo 0x4ac19c <int4pl+12>
>> 0x00000000004ac199 <+9>: cltq
>> 0x00000000004ac19b <+11>: retq
>> 0x00000000004ac19c <+12>: push %rbp
>> 0x00000000004ac19d <+13>: lea 0x1a02c4(%rip),%rsi #
>> 0x64c468
>> 0x00000000004ac1a4 <+20>: xor %r8d,%r8d
>> 0x00000000004ac1a7 <+23>: lea 0x265da1(%rip),%rcx #
>> 0x711f4f <__func__.26823>
>> 0x00000000004ac1ae <+30>: mov $0x30b,%edx
>> 0x00000000004ac1b3 <+35>: mov $0x14,%edi
>> 0x00000000004ac1b8 <+40>: callq 0x586060 <errstart>
>> 0x00000000004ac1bd <+45>: lea 0x147e0e(%rip),%rdi #
>> 0x5f3fd2
>> 0x00000000004ac1c4 <+52>: xor %eax,%eax
>> 0x00000000004ac1c6 <+54>: callq 0x5896a0 <errmsg>
>> 0x00000000004ac1cb <+59>: mov $0x3000082,%edi
>> 0x00000000004ac1d0 <+64>: mov %eax,%ebp
>> 0x00000000004ac1d2 <+66>: callq 0x589540 <errcode>
>> 0x00000000004ac1d7 <+71>: mov %eax,%edi
>> 0x00000000004ac1d9 <+73>: mov %ebp,%esi
>> 0x00000000004ac1db <+75>: xor %eax,%eax
>> 0x00000000004ac1dd <+77>: callq 0x588fb0 <errfinish>
>>
>> out-of-line error:
>> 0x00000000004b04e0 <+0>: mov 0x30(%rdi),%rax
>> 0x00000000004b04e4 <+4>: add 0x20(%rdi),%eax
>> 0x00000000004b04e7 <+7>: jo 0x4b04ec <int4pl+12>
>> 0x00000000004b04e9 <+9>: cltq
>> 0x00000000004b04eb <+11>: retq
>> 0x00000000004b04ec <+12>: push %rax
>> 0x00000000004b04ed <+13>: callq 0x115e17 <out_of_range_err>
>>
>> With the out-of-line error, we can fit multiple of these functions
>into one
>> cache line. With the inline error, not even one.
>>
>> Greetings,
>>
>> Andres Freund
>>
>> [1] https://twitter.com/AndresFreundTec/status/1214305610172289024
>>
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
On Fri, Feb 7, 2020 at 4:54 PM Andres Freund <andres@anarazel.de> wrote:
> On February 6, 2020 11:42:30 PM PST, keisuke kuroda <keisuke.kuroda.3862@gmail.com> wrote:
> >Hi,
> >
> >I have been testing with newer compiler (clang-7)
> >and result is a bit different at least with clang-7.
> >Compiling PG 12.1 (even without patch) with clang-7
> >results in __isinf() no longer being a bottleneck,
> >that is, you don't see it in profiler at all.
>
> I don't think that's necessarily the right conclusion. What's quite possibly happening is that you do not see the
externalisinf function anymore, because it is implemented as an intrinsic, but that there still are more computations
beingdone. Due to the changed order of the isinf checks. You'd have to compare with 11 using the same compiler.
I did some tests using two relatively recent compilers: gcc 8 and
clang-7 and here are the results:
Setup:
create table realtest (a real, b real, c real, d real, e real);
insert into realtest select i, i, i, i, i from generate_series(1, 1000000) i;
Test query:
/tmp/query.sql
select avg(2*dsqrt(a)), avg(2*dsqrt(b)), avg(2*dsqrt(c)),
avg(2*dsqrt(d)), avg(2*dsqrt(e)) from realtest;
pgbench -n -T 60 -f /tmp/query.sql
Latency and profiling results:
gcc 8 (gcc (GCC) 8.3.1 20190311 (Red Hat 8.3.1-3))
====
11.6
latency average = 463.968 ms
40.62% postgres postgres [.] ExecInterpExpr
9.74% postgres postgres [.] float8_accum
6.12% postgres libc-2.17.so [.] __isinf
5.96% postgres postgres [.] float8mul
5.33% postgres postgres [.] dsqrt
3.90% postgres postgres [.] ftod
3.53% postgres postgres [.] Float8GetDatum
2.34% postgres postgres [.] DatumGetFloat8
2.15% postgres postgres [.] AggCheckCallContext
2.03% postgres postgres [.] slot_deform_tuple
1.95% postgres libm-2.17.so [.] __sqrt
1.19% postgres postgres [.] check_float8_array
HEAD
latency average = 549.071 ms
31.74% postgres postgres [.] ExecInterpExpr
11.02% postgres libc-2.17.so [.] __isinf
10.58% postgres postgres [.] float8_accum
4.84% postgres postgres [.] check_float8_val
4.66% postgres postgres [.] dsqrt
3.91% postgres postgres [.] float8mul
3.56% postgres postgres [.] ftod
3.26% postgres postgres [.] Float8GetDatum
2.91% postgres postgres [.] float8_mul
2.30% postgres postgres [.] DatumGetFloat8
2.19% postgres postgres [.] slot_deform_heap_tuple
1.81% postgres postgres [.] AggCheckCallContext
1.31% postgres libm-2.17.so [.] __sqrt
1.25% postgres postgres [.] check_float8_array
HEAD + patch
latency average = 546.624 ms
33.51% postgres postgres [.] ExecInterpExpr
10.35% postgres postgres [.] float8_accum
10.06% postgres libc-2.17.so [.] __isinf
4.58% postgres postgres [.] dsqrt
4.14% postgres postgres [.] check_float8_val
4.03% postgres postgres [.] ftod
3.54% postgres postgres [.] float8mul
2.96% postgres postgres [.] Float8GetDatum
2.38% postgres postgres [.] slot_deform_heap_tuple
2.23% postgres postgres [.] DatumGetFloat8
2.09% postgres postgres [.] float8_mul
1.88% postgres postgres [.] AggCheckCallContext
1.65% postgres libm-2.17.so [.] __sqrt
1.22% postgres postgres [.] check_float8_array
clang-7 (clang version 7.0.1 (tags/RELEASE_701/final))
=====
11.6
latency average = 419.014 ms
47.57% postgres postgres [.] ExecInterpExpr
7.99% postgres postgres [.] float8_accum
5.96% postgres postgres [.] dsqrt
4.88% postgres postgres [.] float8mul
4.23% postgres postgres [.] ftod
3.30% postgres postgres [.] slot_deform_tuple
3.19% postgres postgres [.] DatumGetFloat8
1.92% postgres libm-2.17.so [.] __sqrt
1.72% postgres postgres [.] check_float8_array
HEAD
latency average = 452.958 ms
40.55% postgres postgres [.] ExecInterpExpr
10.61% postgres postgres [.] float8_accum
4.58% postgres postgres [.] dsqrt
3.59% postgres postgres [.] slot_deform_heap_tuple
3.54% postgres postgres [.] check_float8_val
3.48% postgres postgres [.] ftod
3.42% postgres postgres [.] float8mul
3.22% postgres postgres [.] DatumGetFloat8
2.69% postgres postgres [.] Float8GetDatum
2.46% postgres postgres [.] float8_mul
2.29% postgres libm-2.17.so [.] __sqrt
1.47% postgres postgres [.] check_float8_array
HEAD + patch
latency average = 452.533 ms
41.05% postgres postgres [.] ExecInterpExpr
10.15% postgres postgres [.] float8_accum
5.62% postgres postgres [.] dsqrt
3.86% postgres postgres [.] check_float8_val
3.27% postgres postgres [.] float8mul
3.09% postgres postgres [.] slot_deform_heap_tuple
2.91% postgres postgres [.] ftod
2.88% postgres postgres [.] DatumGetFloat8
2.62% postgres postgres [.] float8_mul
2.03% postgres libm-2.17.so [.] __sqrt
2.00% postgres postgres [.] check_float8_array
The patch mentioned above is this:
diff --git a/src/include/utils/float.h b/src/include/utils/float.h
index e2c5dc0f57..dc97d19293 100644
--- a/src/include/utils/float.h
+++ b/src/include/utils/float.h
@@ -136,12 +136,12 @@ static inline void
check_float4_val(const float4 val, const bool inf_is_valid,
const bool zero_is_valid)
{
- if (!inf_is_valid && unlikely(isinf(val)))
+ if (unlikely(isinf(val)) && !inf_is_valid)
ereport(ERROR,
(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
errmsg("value out of range: overflow")));
- if (!zero_is_valid && unlikely(val == 0.0))
+ if (unlikely(val == 0.0) && !zero_is_valid)
ereport(ERROR,
(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
errmsg("value out of range: underflow")));
@@ -151,12 +151,12 @@ static inline void
check_float8_val(const float8 val, const bool inf_is_valid,
const bool zero_is_valid)
{
- if (!inf_is_valid && unlikely(isinf(val)))
+ if (unlikely(isinf(val)) && !inf_is_valid)
ereport(ERROR,
(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
errmsg("value out of range: overflow")));
- if (!zero_is_valid && unlikely(val == 0.0))
+ if (unlikely(val == 0.0) && !zero_is_valid)
ereport(ERROR,
(errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
errmsg("value out of range: underflow")));
So, the patch appears to do very little here. I can only conclude that
the check_float{8|4}_val() (PG 12) is slower than CHECKFLOATVAL() (PG
11) due to arguments being evaluated first. It's entirely possible
though that the patch shown above is not enough.
Thanks,
Amit
Fwiw, also tried the patch that Kuroda-san had posted yesterday.
On Fri, Feb 7, 2020 at 5:17 PM Amit Langote <amitlangote09@gmail.com> wrote:
> Latency and profiling results:
>
> gcc 8 (gcc (GCC) 8.3.1 20190311 (Red Hat 8.3.1-3))
> ====
>
> 11.6
>
> latency average = 463.968 ms
>
> 40.62% postgres postgres [.] ExecInterpExpr
> 9.74% postgres postgres [.] float8_accum
> 6.12% postgres libc-2.17.so [.] __isinf
> 5.96% postgres postgres [.] float8mul
> 5.33% postgres postgres [.] dsqrt
> 3.90% postgres postgres [.] ftod
> 3.53% postgres postgres [.] Float8GetDatum
> 2.34% postgres postgres [.] DatumGetFloat8
> 2.15% postgres postgres [.] AggCheckCallContext
> 2.03% postgres postgres [.] slot_deform_tuple
> 1.95% postgres libm-2.17.so [.] __sqrt
> 1.19% postgres postgres [.] check_float8_array
>
> HEAD
>
> latency average = 549.071 ms
>
> 31.74% postgres postgres [.] ExecInterpExpr
> 11.02% postgres libc-2.17.so [.] __isinf
> 10.58% postgres postgres [.] float8_accum
> 4.84% postgres postgres [.] check_float8_val
> 4.66% postgres postgres [.] dsqrt
> 3.91% postgres postgres [.] float8mul
> 3.56% postgres postgres [.] ftod
> 3.26% postgres postgres [.] Float8GetDatum
> 2.91% postgres postgres [.] float8_mul
> 2.30% postgres postgres [.] DatumGetFloat8
> 2.19% postgres postgres [.] slot_deform_heap_tuple
> 1.81% postgres postgres [.] AggCheckCallContext
> 1.31% postgres libm-2.17.so [.] __sqrt
> 1.25% postgres postgres [.] check_float8_array
>
> HEAD + patch
>
> latency average = 546.624 ms
>
> 33.51% postgres postgres [.] ExecInterpExpr
> 10.35% postgres postgres [.] float8_accum
> 10.06% postgres libc-2.17.so [.] __isinf
> 4.58% postgres postgres [.] dsqrt
> 4.14% postgres postgres [.] check_float8_val
> 4.03% postgres postgres [.] ftod
> 3.54% postgres postgres [.] float8mul
> 2.96% postgres postgres [.] Float8GetDatum
> 2.38% postgres postgres [.] slot_deform_heap_tuple
> 2.23% postgres postgres [.] DatumGetFloat8
> 2.09% postgres postgres [.] float8_mul
> 1.88% postgres postgres [.] AggCheckCallContext
> 1.65% postgres libm-2.17.so [.] __sqrt
> 1.22% postgres postgres [.] check_float8_array
HEAD + Kuroda-san's patch (compiled with gcc 8)
latency average = 484.604 ms
37.41% postgres postgres [.] ExecInterpExpr
10.83% postgres postgres [.] float8_accum
5.62% postgres postgres [.] dsqrt
4.23% postgres libc-2.17.so [.] __isinf
4.05% postgres postgres [.] float8mul
3.85% postgres postgres [.] ftod
3.18% postgres postgres [.] Float8GetDatum
2.81% postgres postgres [.] slot_deform_heap_tuple
2.63% postgres postgres [.] DatumGetFloat8
2.46% postgres postgres [.] float8_mul
1.91% postgres libm-2.17.so [.] __sqrt
> clang-7 (clang version 7.0.1 (tags/RELEASE_701/final))
> =====
>
> 11.6
>
> latency average = 419.014 ms
>
> 47.57% postgres postgres [.] ExecInterpExpr
> 7.99% postgres postgres [.] float8_accum
> 5.96% postgres postgres [.] dsqrt
> 4.88% postgres postgres [.] float8mul
> 4.23% postgres postgres [.] ftod
> 3.30% postgres postgres [.] slot_deform_tuple
> 3.19% postgres postgres [.] DatumGetFloat8
> 1.92% postgres libm-2.17.so [.] __sqrt
> 1.72% postgres postgres [.] check_float8_array
>
> HEAD
>
> latency average = 452.958 ms
>
> 40.55% postgres postgres [.] ExecInterpExpr
> 10.61% postgres postgres [.] float8_accum
> 4.58% postgres postgres [.] dsqrt
> 3.59% postgres postgres [.] slot_deform_heap_tuple
> 3.54% postgres postgres [.] check_float8_val
> 3.48% postgres postgres [.] ftod
> 3.42% postgres postgres [.] float8mul
> 3.22% postgres postgres [.] DatumGetFloat8
> 2.69% postgres postgres [.] Float8GetDatum
> 2.46% postgres postgres [.] float8_mul
> 2.29% postgres libm-2.17.so [.] __sqrt
> 1.47% postgres postgres [.] check_float8_array
>
> HEAD + patch
>
> latency average = 452.533 ms
>
> 41.05% postgres postgres [.] ExecInterpExpr
> 10.15% postgres postgres [.] float8_accum
> 5.62% postgres postgres [.] dsqrt
> 3.86% postgres postgres [.] check_float8_val
> 3.27% postgres postgres [.] float8mul
> 3.09% postgres postgres [.] slot_deform_heap_tuple
> 2.91% postgres postgres [.] ftod
> 2.88% postgres postgres [.] DatumGetFloat8
> 2.62% postgres postgres [.] float8_mul
> 2.03% postgres libm-2.17.so [.] __sqrt
> 2.00% postgres postgres [.] check_float8_array
HEAD + Kuroda-san's patch (compiled with clang-7)
latency average = 435.454 ms
43.02% postgres postgres [.] ExecInterpExpr
10.86% postgres postgres [.] float8_accum
3.97% postgres postgres [.] dsqrt
3.97% postgres postgres [.] float8mul
3.51% postgres postgres [.] ftod
3.42% postgres postgres [.] slot_deform_heap_tuple
3.36% postgres postgres [.] DatumGetFloat8
1.97% postgres libm-2.17.so [.] __sqrt
1.97% postgres postgres [.] check_float8_array
1.88% postgres postgres [.] float8_mul
Needless to say, that one makes a visible difference, although still
slower compared to PG 11.
Thanks,
Amit
> Fwiw, also tried the patch that Kuroda-san had posted yesterday. I run the same test case too: clang version 7.0.0: HEAD 2548.119 ms with patch 2320.974 ms clang version 8.0.0: HEAD 2431.766 ms with patch 2419.439 ms clang version 9.0.0: HEAD 2477.493 ms with patch 2365.509 ms gcc version 7.4.0: HEAD 2451.261 ms with patch 2343.393 ms gcc version 8.3.0: HEAD 2540.626 ms with patch 2299.653 ms
> > The patch looks unduly invasive to me, but I think that it might be > > right that we should go back to a macro-based implementation, because > > otherwise we don't have a good way to be certain that the function > > parameter won't get evaluated first. > > I'd first like to see some actual evidence of this being a problem, > rather than just the order of the checks. There seem to be enough evidence of this being the problem. We are better off going back to the macro-based implementation. I polished Keisuke Kuroda's patch commenting about the performance issue, removed the check_float*_val() functions completely, and added unlikely() as Tom Lane suggested. It is attached. I confirmed with different compilers that the macro, and unlikely() makes this noticeably faster.
Вложения
Moin, On 2020-02-07 15:42, Emre Hasegeli wrote: >> > The patch looks unduly invasive to me, but I think that it might be >> > right that we should go back to a macro-based implementation, because >> > otherwise we don't have a good way to be certain that the function >> > parameter won't get evaluated first. >> >> I'd first like to see some actual evidence of this being a problem, >> rather than just the order of the checks. > > There seem to be enough evidence of this being the problem. We are > better off going back to the macro-based implementation. I polished > Keisuke Kuroda's patch commenting about the performance issue, removed > the check_float*_val() functions completely, and added unlikely() as > Tom Lane suggested. It is attached. I confirmed with different > compilers that the macro, and unlikely() makes this noticeably faster. Hm, the diff has the macro tests as: + if (unlikely(isinf(val) && !(inf_is_valid))) ... + if (unlikely((val) == 0.0 && !(zero_is_valid))) But the comment does not explain that this test has to be in that order, or the compiler will for non-constant arguments evalute the (now) right-side first. E.g. if I understand this correctly: + if (!(zero_is_valid) && unlikely((val) == 0.0) would have the same problem of evaluating "zero_is_valid" (which might be an isinf(arg1) || isinf(arg2)) first and so be the same thing we try to avoid with the macro? Maybe adding this bit of info to the comment makes it clearer? Also, a few places use the macro as: + CHECKFLOATVAL(result, true, true); which evaluates to a complete NOP in both cases. IMHO this could be replaced with a comment like: + // No CHECKFLOATVAL() needed, as both inf and 0.0 are valid (or something along the lines of "no error can occur"), as otherwise CHECKFLOATVAL() implies to the casual reader that there are some checks done, while in reality no real checks are done at all (and hopefully the compiler optimizes everything away, which might not be true for debug builds). -- Best regards, Tels
Вложения
Hi, On 2020-02-07 17:17:21 +0900, Amit Langote wrote: > I did some tests using two relatively recent compilers: gcc 8 and > clang-7 and here are the results: Hm, these very much look like they've been done in an unoptimized build? > 40.62% postgres postgres [.] ExecInterpExpr > 9.74% postgres postgres [.] float8_accum > 6.12% postgres libc-2.17.so [.] __isinf > 5.96% postgres postgres [.] float8mul > 5.33% postgres postgres [.] dsqrt > 3.90% postgres postgres [.] ftod > 3.53% postgres postgres [.] Float8GetDatum > 2.34% postgres postgres [.] DatumGetFloat8 > 2.15% postgres postgres [.] AggCheckCallContext > 2.03% postgres postgres [.] slot_deform_tuple > 1.95% postgres libm-2.17.so [.] __sqrt > 1.19% postgres postgres [.] check_float8_array > HEAD > > latency average = 549.071 ms > > 31.74% postgres postgres [.] ExecInterpExpr > 11.02% postgres libc-2.17.so [.] __isinf > 10.58% postgres postgres [.] float8_accum > 4.84% postgres postgres [.] check_float8_val > 4.66% postgres postgres [.] dsqrt > 3.91% postgres postgres [.] float8mul > 3.56% postgres postgres [.] ftod > 3.26% postgres postgres [.] Float8GetDatum > 2.91% postgres postgres [.] float8_mul > 2.30% postgres postgres [.] DatumGetFloat8 > 2.19% postgres postgres [.] slot_deform_heap_tuple > 1.81% postgres postgres [.] AggCheckCallContext > 1.31% postgres libm-2.17.so [.] __sqrt > 1.25% postgres postgres [.] check_float8_array Because DatumGetFloat8, Float8GetDatum, etc aren't functions that normally stay separate. Greetings, Andres Freund
On Sat, Feb 8, 2020 at 3:13 AM Andres Freund <andres@anarazel.de> wrote:
> On 2020-02-07 17:17:21 +0900, Amit Langote wrote:
> > I did some tests using two relatively recent compilers: gcc 8 and
> > clang-7 and here are the results:
>
> Hm, these very much look like they've been done in an unoptimized build?
>
> > 40.62% postgres postgres [.] ExecInterpExpr
> > 9.74% postgres postgres [.] float8_accum
> > 6.12% postgres libc-2.17.so [.] __isinf
> > 5.96% postgres postgres [.] float8mul
> > 5.33% postgres postgres [.] dsqrt
> > 3.90% postgres postgres [.] ftod
> > 3.53% postgres postgres [.] Float8GetDatum
> > 2.34% postgres postgres [.] DatumGetFloat8
> > 2.15% postgres postgres [.] AggCheckCallContext
> > 2.03% postgres postgres [.] slot_deform_tuple
> > 1.95% postgres libm-2.17.so [.] __sqrt
> > 1.19% postgres postgres [.] check_float8_array
>
> > HEAD
> >
> > latency average = 549.071 ms
> >
> > 31.74% postgres postgres [.] ExecInterpExpr
> > 11.02% postgres libc-2.17.so [.] __isinf
> > 10.58% postgres postgres [.] float8_accum
> > 4.84% postgres postgres [.] check_float8_val
> > 4.66% postgres postgres [.] dsqrt
> > 3.91% postgres postgres [.] float8mul
> > 3.56% postgres postgres [.] ftod
> > 3.26% postgres postgres [.] Float8GetDatum
> > 2.91% postgres postgres [.] float8_mul
> > 2.30% postgres postgres [.] DatumGetFloat8
> > 2.19% postgres postgres [.] slot_deform_heap_tuple
> > 1.81% postgres postgres [.] AggCheckCallContext
> > 1.31% postgres libm-2.17.so [.] __sqrt
> > 1.25% postgres postgres [.] check_float8_array
>
> Because DatumGetFloat8, Float8GetDatum, etc aren't functions that
> normally stay separate.
Okay, fair.
Here are numbers after compiling with -O3:
gcc 8
=====
HEAD
latency average = 350.187 ms
34.67% postgres postgres [.] ExecInterpExpr
20.94% postgres libc-2.17.so [.] __isinf
10.74% postgres postgres [.] float8_accum
8.22% postgres postgres [.] dsqrt
6.63% postgres postgres [.] float8mul
3.45% postgres postgres [.] ftod
2.32% postgres postgres [.] tts_buffer_heap_getsomeattrs
HEAD + reverse-if-condition patch
latency average = 346.710 ms
34.48% postgres postgres [.] ExecInterpExpr
21.00% postgres libc-2.17.so [.] __isinf
12.26% postgres postgres [.] float8_accum
8.31% postgres postgres [.] dsqrt
6.32% postgres postgres [.] float8mul
3.23% postgres postgres [.] ftod
2.25% postgres postgres [.] tts_buffer_heap_getsomeattrs
HEAD + revert-to-macro patch
latency average = 297.493 ms
39.25% postgres postgres [.] ExecInterpExpr
14.44% postgres postgres [.] float8_accum
11.02% postgres libc-2.17.so [.] __isinf
8.21% postgres postgres [.] dsqrt
5.55% postgres postgres [.] float8mul
4.15% postgres postgres [.] ftod
2.78% postgres postgres [.] tts_buffer_heap_getsomeattrs
11.6
latency average = 290.301 ms
42.78% postgres postgres [.] ExecInterpExpr
12.27% postgres postgres [.] float8_accum
12.12% postgres libc-2.17.so [.] __isinf
8.96% postgres postgres [.] dsqrt
5.77% postgres postgres [.] float8mul
3.94% postgres postgres [.] ftod
2.61% postgres postgres [.] AggCheckCallContext
clang-7
=======
HEAD
latency average = 246.278 ms
44.47% postgres postgres [.] ExecInterpExpr
14.56% postgres postgres [.] float8_accum
7.25% postgres postgres [.] float8mul
7.22% postgres postgres [.] dsqrt
5.40% postgres postgres [.] ftod
4.09% postgres postgres [.] tts_buffer_heap_getsomeattrs
2.20% postgres postgres [.] check_float8_val
HEAD + reverse-if-condition patch
latency average = 240.212 ms
45.49% postgres postgres [.] ExecInterpExpr
13.69% postgres postgres [.] float8_accum
8.32% postgres postgres [.] dsqrt
5.28% postgres postgres [.] ftod
5.19% postgres postgres [.] float8mul
3.68% postgres postgres [.] tts_buffer_heap_getsomeattrs
2.90% postgres postgres [.] float8_mul
HEAD + revert-to-macro patch
latency average = 240.620 ms
44.04% postgres postgres [.] ExecInterpExpr
13.72% postgres postgres [.] float8_accum
9.26% postgres postgres [.] dsqrt
5.30% postgres postgres [.] ftod
4.66% postgres postgres [.] float8mul
3.53% postgres postgres [.] tts_buffer_heap_getsomeattrs
3.39% postgres postgres [.] float8_mul
11.6
latency average = 237.045 ms
46.85% postgres postgres [.] ExecInterpExpr
11.39% postgres postgres [.] float8_accum
8.02% postgres postgres [.] dsqrt
7.29% postgres postgres [.] slot_deform_tuple
6.04% postgres postgres [.] float8mul
5.49% postgres postgres [.] ftod
PG 12 is worse than PG 11 when compiled with gcc.
Thanks,
Amit
On Fri, Feb 7, 2020 at 11:43 PM Emre Hasegeli <emre@hasegeli.com> wrote: > > > The patch looks unduly invasive to me, but I think that it might be > > > right that we should go back to a macro-based implementation, because > > > otherwise we don't have a good way to be certain that the function > > > parameter won't get evaluated first. > > > > I'd first like to see some actual evidence of this being a problem, > > rather than just the order of the checks. > > There seem to be enough evidence of this being the problem. We are > better off going back to the macro-based implementation. I polished > Keisuke Kuroda's patch commenting about the performance issue, removed > the check_float*_val() functions completely, and added unlikely() as > Tom Lane suggested. It is attached. I confirmed with different > compilers that the macro, and unlikely() makes this noticeably faster. Thanks for updating the patch. Should we update the same macro in contrib/btree_gist/btree_utils_num.h too? Regards, Amit
> But the comment does not explain that this test has to be in that > order, or the compiler will for non-constant arguments evalute > the (now) right-side first. E.g. if I understand this correctly: > > + if (!(zero_is_valid) && unlikely((val) == 0.0) > > would have the same problem of evaluating "zero_is_valid" (which > might be an isinf(arg1) || isinf(arg2)) first and so be the same thing > we try to avoid with the macro? Maybe adding this bit of info to the > comment makes it clearer? Added. > Also, a few places use the macro as: > > + CHECKFLOATVAL(result, true, true); > > which evaluates to a complete NOP in both cases. IMHO this could be > replaced with a comment like: > > + // No CHECKFLOATVAL() needed, as both inf and 0.0 are valid > > (or something along the lines of "no error can occur"), as otherwise > CHECKFLOATVAL() implies to the casual reader that there are some checks > done, while in reality no real checks are done at all (and hopefully > the compiler optimizes everything away, which might not be true for > debug builds). I don't know why those trigonometric functions don't check for overflow/underflow like all the rest of float.c. I'll submit another patch to make them error when overflow/underflow. The new version is attached.
Вложения
> Should we update the same macro in contrib/btree_gist/btree_utils_num.h too? I posted another version incorporating this.
Hi, On 2020-02-12 11:54:13 +0000, Emre Hasegeli wrote: > From fb5052b869255ef9465b1de92e84b2fb66dd6eb3 Mon Sep 17 00:00:00 2001 > From: Emre Hasegeli <emre@hasegeli.com> > Date: Fri, 7 Feb 2020 10:27:25 +0000 > Subject: [PATCH] Bring back CHECKFLOATVAL() macro > > The inline functions added by 6bf0bc842b caused the conditions of > overflow/underflow checks to be evaluated when no overflow/underflow > happen. This slowed down floating point operations. This commit brings > back the macro that was in use before 6bf0bc842b to fix the performace > regression. Wait, no. Didn't we get to the point that we figured out that the primary issue is the reversal of the order of what is checked is the primary problem, rather than the macro/inline piece? Nor do I see how it's going to be ok to just rename the function in a stable branch. Greetings, Andres Freund
> Wait, no. Didn't we get to the point that we figured out that the > primary issue is the reversal of the order of what is checked is the > primary problem, rather than the macro/inline piece? Reversal of the order makes a small or no difference. The macro/inline change causes the real slowdown at least on GCC. > Nor do I see how it's going to be ok to just rename the function in a > stable branch. I'll post another version to keep them around.
On 2020-02-12 17:49:14 +0000, Emre Hasegeli wrote: > > Nor do I see how it's going to be ok to just rename the function in a > > stable branch. > > I'll post another version to keep them around. I'd just rename the macro to the name of the inline function. No need to have a verbose change in all callsites just to update the name imo.
Andres Freund <andres@anarazel.de> writes:
> I'd just rename the macro to the name of the inline function. No need to
> have a verbose change in all callsites just to update the name imo.
+1, that's what I had in mind too. That does suggest though that we
ought to make sure the macro has single-eval behavior, so that you
don't need to know it's a macro.
regards, tom lane
Hi,
On 2020-02-12 13:15:22 -0500, Tom Lane wrote:
> Andres Freund <andres@anarazel.de> writes:
> > I'd just rename the macro to the name of the inline function. No need to
> > have a verbose change in all callsites just to update the name imo.
>
> +1, that's what I had in mind too. That does suggest though that we
> ought to make sure the macro has single-eval behavior, so that you
> don't need to know it's a macro.
We'd have to store 'val' in a local variable for that I think. Not the
prettiest, but also not a problem.
I do wonder if we're just punching ourselves in the face with the
signature of these checks. Part of the problem here really comes from
using the same function to handle a number of different checks.
I mean something like dtof's
check_float4_val((float4) num, isinf(num), num == 0);
where the num == 0 is solely to satisfy the check function is a bit
stupid.
And the reason we have these isinf(arg1) || isinf(arg2) parameters is
also largely because we force the same function to be used in cases
where we have two inputs, rather than just one.
For most places it'd probably end up being easier to read and to
optimize if we just wrote them as
if (unlikely(isinf(result)) && !isinf(arg))
float_overflow_error();
and when needed added a
else if (unlikely(result == 0) && arg1 != 0.0)
float_underflow_error();
the verbose piece really is the error, not the error check. Sure, there
are more complicated cases like
if (unlikely(isinf(result)) && (!isinf(arg1) || !isinf(arg2)))
but that's still not very complicated.
Greetings,
Andres Freund
Andres Freund <andres@anarazel.de> writes:
> I do wonder if we're just punching ourselves in the face with the
> signature of these checks. Part of the problem here really comes from
> using the same function to handle a number of different checks.
Yeah, I've thought that too. It's *far* from clear that this thing
is a win at all, other than your point about the number of copies of
the ereport call. It's bulky, it's hard to optimize, and I have
never thought it was more readable than the direct tests it replaced.
> For most places it'd probably end up being easier to read and to
> optimize if we just wrote them as
> if (unlikely(isinf(result)) && !isinf(arg))
> float_overflow_error();
> and when needed added a
> else if (unlikely(result == 0) && arg1 != 0.0)
> float_underflow_error();
+1
regards, tom lane
Hi, On 2020-02-12 14:18:30 -0500, Tom Lane wrote: > Andres Freund <andres@anarazel.de> writes: > > I do wonder if we're just punching ourselves in the face with the > > signature of these checks. Part of the problem here really comes from > > using the same function to handle a number of different checks. > > Yeah, I've thought that too. It's *far* from clear that this thing > is a win at all, other than your point about the number of copies of > the ereport call. It's bulky, it's hard to optimize, and I have > never thought it was more readable than the direct tests it replaced. > > > For most places it'd probably end up being easier to read and to > > optimize if we just wrote them as > > if (unlikely(isinf(result)) && !isinf(arg)) > > float_overflow_error(); > > and when needed added a > > else if (unlikely(result == 0) && arg1 != 0.0) > > float_underflow_error(); > > +1 Cool. Emre, any chance you could write a patch along those lines? I'm inclined that we should backpatch that, and just leave the inline function (without in core callers) in place in 12? Greetings, Andres Freund
Andres Freund <andres@anarazel.de> writes:
> I'm inclined that we should backpatch that, and just leave the inline
> function (without in core callers) in place in 12?
Yeah, we can't remove the inline function in 12. But we don't have
to use it.
regards, tom lane
> > > For most places it'd probably end up being easier to read and to > > > optimize if we just wrote them as > > > if (unlikely(isinf(result)) && !isinf(arg)) > > > float_overflow_error(); > > > and when needed added a > > > else if (unlikely(result == 0) && arg1 != 0.0) > > > float_underflow_error(); > > > > +1 > > Cool. Emre, any chance you could write a patch along those lines? Yes, I am happy to do. It makes more sense to me too.
> > > > For most places it'd probably end up being easier to read and to > > > > optimize if we just wrote them as > > > > if (unlikely(isinf(result)) && !isinf(arg)) > > > > float_overflow_error(); > > > > and when needed added a > > > > else if (unlikely(result == 0) && arg1 != 0.0) > > > > float_underflow_error(); > > > > > > +1 > > > > Cool. Emre, any chance you could write a patch along those lines? > > Yes, I am happy to do. It makes more sense to me too. How about the one attached?
Вложения
Emre Hasegeli <emre@hasegeli.com> writes:
>>> Cool. Emre, any chance you could write a patch along those lines?
> How about the one attached?
I see some minor things I don't like here, eg float_*flow_error()
need some documentation as to why they exist. But I'll review,
fix those things up and then push.
regards, tom lane
Hi, On February 13, 2020 8:30:45 AM PST, Tom Lane <tgl@sss.pgh.pa.us> wrote: >Emre Hasegeli <emre@hasegeli.com> writes: >>>> Cool. Emre, any chance you could write a patch along those lines? > >> How about the one attached? > >I see some minor things I don't like here, eg float_*flow_error() >need some documentation as to why they exist. But I'll review, >fix those things up and then push. Would be good to mark them noreturn too. Wonder if it's useful to add the"cold" marker to pg. Not as part of this patch, but for functions like these. Andres -- Sent from my Android device with K-9 Mail. Please excuse my brevity.
Hi,
On 2020-02-13 16:25:25 +0000, Emre Hasegeli wrote:
> And also this commit is changing the usage of unlikely() to cover
> the whole condition. Using it only for the result is not semantically
> correct. It is more than likely for the result to be infinite when
> the input is, or it to be 0 when the input is.
I'm not really convinced by this fwiw.
Comparing
if (unlikely(isinf(result) && !isinf(num)))
float_overflow_error();
with
if (unlikely(isinf(result)) && !isinf(num))
float_overflow_error();
I don't think it's clear that we want the former. What we want to
express is that it's unlikely that the result is infinite, and that the
compiler should optimize for that. Since there's a jump involved between
the check for isinf(result) and the one for !isinf(num), we want the
compiler to implement this so the non-overflow path follows the first
check, and the rest of the check is later.
> +void float_overflow_error()
> +{
Tom's probably on this, but it should be (void).
> @@ -2846,23 +2909,21 @@ float8_accum(PG_FUNCTION_ARGS)
>
> /*
> * Overflow check. We only report an overflow error when finite
> * inputs lead to infinite results. Note also that Sxx should be NaN
> * if any of the inputs are infinite, so we intentionally prevent Sxx
> * from becoming infinite.
> */
> if (isinf(Sx) || isinf(Sxx))
> {
> if (!isinf(transvalues[1]) && !isinf(newval))
> - ereport(ERROR,
> - (errcode(ERRCODE_NUMERIC_VALUE_OUT_OF_RANGE),
> - errmsg("value out of range: overflow")));
> + float_overflow_error();
>
> Sxx = get_float8_nan();
> }
> }
Probably worth unifiying the use of unlikely around isinf here and in
the follow functions.
Greetings,
Andres Freund
Andres Freund <andres@anarazel.de> writes:
> On February 13, 2020 8:30:45 AM PST, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> I see some minor things I don't like here, eg float_*flow_error()
>> need some documentation as to why they exist. But I'll review,
>> fix those things up and then push.
> Would be good to mark them noreturn too.
Yeah, that was one of the things I didn't like ;-). Also the lack
of pg_noinline.
> Wonder if it's useful to add the"cold" marker to pg. Not as part of this patch, but for functions like these.
I'm only seeing about a 1.5kB reduction in the backend size from
this patch, which kinda surprises me, but it says that we're
not winning all that much from just having one copy of the ereport
calls. So I don't think that "cold" is going to add much.
regards, tom lane
Andres Freund <andres@anarazel.de> writes:
> On 2020-02-13 16:25:25 +0000, Emre Hasegeli wrote:
>> And also this commit is changing the usage of unlikely() to cover
>> the whole condition. Using it only for the result is not semantically
>> correct. It is more than likely for the result to be infinite when
>> the input is, or it to be 0 when the input is.
> I'm not really convinced by this fwiw.
> Comparing
> if (unlikely(isinf(result) && !isinf(num)))
> float_overflow_error();
> with
> if (unlikely(isinf(result)) && !isinf(num))
> float_overflow_error();
> I don't think it's clear that we want the former. What we want to
> express is that it's unlikely that the result is infinite, and that the
> compiler should optimize for that. Since there's a jump involved between
> the check for isinf(result) and the one for !isinf(num), we want the
> compiler to implement this so the non-overflow path follows the first
> check, and the rest of the check is later.
Yeah, I was wondering about that. I'll change it as you suggest.
regards, tom lane
... and pushed. One other change I made beyond those suggested
was to push the zero-divide ereport's out-of-line as well.
I did not do anything about adding unlikely() calls around the
unrelated isinf tests in float.c. That seemed to me to be a separate
matter, and I'm not quite convinced it'd be a win anyway.
regards, tom lane
Hi, On 2020-02-13 13:40:43 -0500, Tom Lane wrote: > ... and pushed. One other change I made beyond those suggested > was to push the zero-divide ereport's out-of-line as well. Thanks! > I did not do anything about adding unlikely() calls around the > unrelated isinf tests in float.c. That seemed to me to be a separate > matter, and I'm not quite convinced it'd be a win anyway. I was mostly going for consistency... Greetings, Andres Freund
On Fri, Feb 14, 2020 at 3:47 AM Andres Freund <andres@anarazel.de> wrote:
> On 2020-02-13 13:40:43 -0500, Tom Lane wrote:
> > ... and pushed. One other change I made beyond those suggested
> > was to push the zero-divide ereport's out-of-line as well.
>
> Thanks!
Thank you all.
I repeated some of the tests I did earlier and things look good.
gcc-8
=====
HEAD
latency average = 296.842 ms
42.05% postgres postgres [.] ExecInterpExpr
15.14% postgres postgres [.] float8_accum
9.32% postgres libc-2.17.so [.] __isinf
7.32% postgres postgres [.] dsqrt
5.67% postgres postgres [.] float8mul
4.20% postgres postgres [.] ftod
11.7
latency average = 289.439 ms
41.52% postgres postgres [.] ExecInterpExpr
13.59% postgres libc-2.17.so [.] __isinf
10.98% postgres postgres [.] float8_accum
8.26% postgres postgres [.] dsqrt
6.17% postgres postgres [.] float8mul
3.65% postgres postgres [.] ftod
clang-7
=======
HEAD
latency average = 233.735 ms
43.84% postgres postgres [.] ExecInterpExpr
15.17% postgres postgres [.] float8_accum
8.25% postgres postgres [.] dsqrt
7.35% postgres postgres [.] float8mul
5.84% postgres postgres [.] ftod
3.78% postgres postgres [.] tts_buffer_heap_getsomeattrs
11.7
latency average = 221.009 ms
49.55% postgres postgres [.] ExecInterpExpr
12.05% postgres postgres [.] float8_accum
8.97% postgres postgres [.] dsqrt
6.72% postgres postgres [.] float8mul
5.62% postgres postgres [.] ftod
2.18% postgres postgres [.] slot_deform_tuple
HEAD and PG 11 are now comparable even when built with gcc.
Regards,
Amit
Thank you very much everyone.
Improvement was confirmed even if PG12_STABLE was built with gcc 4.8.5.
* PG_12_STABLE
* gcc 4.8.5
postgres=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------
Seq Scan on public.realtest (cost=0.00..288692.14 rows=9999873 width=40) (actual time=0.012..4118.432 rows=10000001 loops=1)
Output: ('2'::double precision * a), ('2'::double precision * b), ('2'::double precision * c), ('2'::double precision * d), ('2'::double precision * e)
Buffers: shared hit=63695
Planning Time: 0.034 ms
Execution Time: 4811.957 ms
(5 rows)
32.03% postgres postgres [.] ExecInterpExpr
12.28% postgres postgres [.] float84mul
9.62% postgres [vdso] [.] __vdso_clock_gettime
6.45% postgres libc-2.17.so [.] __isinf
5.15% postgres postgres [.] tts_buffer_heap_getsomeattrs
3.83% postgres postgres [.] ExecScan
Best Regards,
Keisuke Kuroda
Improvement was confirmed even if PG12_STABLE was built with gcc 4.8.5.
* PG_12_STABLE
* gcc 4.8.5
postgres=# EXPLAIN (ANALYZE on, VERBOSE on, BUFFERS on)
select (2 * a) , (2 * b) , (2 * c), (2 * d), (2 * e)
from realtest;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------
Seq Scan on public.realtest (cost=0.00..288692.14 rows=9999873 width=40) (actual time=0.012..4118.432 rows=10000001 loops=1)
Output: ('2'::double precision * a), ('2'::double precision * b), ('2'::double precision * c), ('2'::double precision * d), ('2'::double precision * e)
Buffers: shared hit=63695
Planning Time: 0.034 ms
Execution Time: 4811.957 ms
(5 rows)
32.03% postgres postgres [.] ExecInterpExpr
12.28% postgres postgres [.] float84mul
9.62% postgres [vdso] [.] __vdso_clock_gettime
6.45% postgres libc-2.17.so [.] __isinf
5.15% postgres postgres [.] tts_buffer_heap_getsomeattrs
3.83% postgres postgres [.] ExecScan
Best Regards,
Keisuke Kuroda
2020年2月14日(金) 13:29 Amit Langote <amitlangote09@gmail.com>:
On Fri, Feb 14, 2020 at 3:47 AM Andres Freund <andres@anarazel.de> wrote:
> On 2020-02-13 13:40:43 -0500, Tom Lane wrote:
> > ... and pushed. One other change I made beyond those suggested
> > was to push the zero-divide ereport's out-of-line as well.
>
> Thanks!
Thank you all.
I repeated some of the tests I did earlier and things look good.
gcc-8
=====
HEAD
latency average = 296.842 ms
42.05% postgres postgres [.] ExecInterpExpr
15.14% postgres postgres [.] float8_accum
9.32% postgres libc-2.17.so [.] __isinf
7.32% postgres postgres [.] dsqrt
5.67% postgres postgres [.] float8mul
4.20% postgres postgres [.] ftod
11.7
latency average = 289.439 ms
41.52% postgres postgres [.] ExecInterpExpr
13.59% postgres libc-2.17.so [.] __isinf
10.98% postgres postgres [.] float8_accum
8.26% postgres postgres [.] dsqrt
6.17% postgres postgres [.] float8mul
3.65% postgres postgres [.] ftod
clang-7
=======
HEAD
latency average = 233.735 ms
43.84% postgres postgres [.] ExecInterpExpr
15.17% postgres postgres [.] float8_accum
8.25% postgres postgres [.] dsqrt
7.35% postgres postgres [.] float8mul
5.84% postgres postgres [.] ftod
3.78% postgres postgres [.] tts_buffer_heap_getsomeattrs
11.7
latency average = 221.009 ms
49.55% postgres postgres [.] ExecInterpExpr
12.05% postgres postgres [.] float8_accum
8.97% postgres postgres [.] dsqrt
6.72% postgres postgres [.] float8mul
5.62% postgres postgres [.] ftod
2.18% postgres postgres [.] slot_deform_tuple
HEAD and PG 11 are now comparable even when built with gcc.
Regards,
Amit