From: Vladislav Shpilevoy <v.shpilevoy@tarantool.org>
To: tarantool-patches@dev.tarantool.org, kyukhin@tarantool.org
Subject: [Tarantool-patches] [PATCH 1/1] tuple: don't truncate float in :update()
Date: Sun, 23 Feb 2020 15:11:31 +0100 [thread overview]
Message-ID: <2b1afed1b05757c177b2fbb2efbf7643670c43cc.1582466958.git.v.shpilevoy@tarantool.org> (raw)
Before the patch there were the rules:
* float +/- double = double
* double +/- double = double
* float +/- float = float
The rules were applied regardless of values. That led to a problem
when float + float exceeding maximal float value could fit into
double, but was stored as an infinity.
The patch makes so that if a floating point arithmetic operation
result fits into float, it is stored as float. Otherwise as
double. Regardless of initial types.
This alongside saves some memory for cases when doubles can be
stored as floats, and therefore takes 4 less bytes. Although
these cases are rare, because any not integer value stored in a
double may have a long garbage tail in its fraction.
Closes #4701
(cherry picked from commit fef4fdfc3edee39c72f2e93e6f96ab681750617f)
---
Branch: https://github.com/tarantool/tarantool/tree/gerold103/gh-4701-update-float-2.2
Issue: https://github.com/tarantool/tarantool/issues/4701
Fix for 4701 was not pushed to 2.2 and 1.10. But it is a bug fix,
and it should be pushed to them too. This branch is for 2.2 and
it should be easy to cherry-pick it to 1.10.
@ChangeLog
- tuple/space/index:update()/upsert() were fixed not to turn a
value into an infinity, when a float value was added to or
subtracted from a float column and exceeded float value range
(gh-4701).
src/box/tuple_update.c | 11 +++----
test/box/update.result | 65 ++++++++++++++++++++++++++++++++++++++++
test/box/update.test.lua | 45 ++++++++++++++++++++++++++++
3 files changed, 116 insertions(+), 5 deletions(-)
diff --git a/src/box/tuple_update.c b/src/box/tuple_update.c
index 7a203ced8..785a88640 100644
--- a/src/box/tuple_update.c
+++ b/src/box/tuple_update.c
@@ -33,6 +33,7 @@
#include <stdio.h>
#include <stdbool.h>
+#include <float.h>
#include "say.h"
#include "error.h"
@@ -484,15 +485,15 @@ make_arith_operation(struct op_arith_arg arg1, struct op_arith_arg arg2,
err_fieldno, "a positive integer");
return -1;
}
- if (lowest_type == AT_DOUBLE) {
- /* result is DOUBLE */
+ float fc = (float)c;
+ if ((lowest_type == AT_DOUBLE && c != (double) fc) ||
+ (lowest_type == AT_FLOAT &&
+ (c >= FLT_MAX || c <= -FLT_MAX))) {
ret->type = AT_DOUBLE;
ret->dbl = c;
} else {
- /* result is FLOAT */
- assert(lowest_type == AT_FLOAT);
ret->type = AT_FLOAT;
- ret->flt = (float)c;
+ ret->flt = fc;
}
}
return 0;
diff --git a/test/box/update.result b/test/box/update.result
index a3f731b55..0dbe0f0a1 100644
--- a/test/box/update.result
+++ b/test/box/update.result
@@ -849,3 +849,68 @@ s:update(1, {{'=', 3, map}})
s:drop()
---
...
+--
+-- gh-4701: arith operations should not truncate result when
+-- float + float fits double.
+--
+msgpackffi = require('msgpackffi')
+---
+...
+mp_array_1 = 0x91
+---
+...
+mp_double = 0xcb
+---
+...
+mp_float = 0xca
+---
+...
+flt_max = 3.402823466e+38
+---
+...
+uint_max = 18446744073709551615ULL
+---
+...
+-- Double + double is float if result fits. \
+-- Double + double is double if result does not fit float. \
+-- Double + float is float if result fits. \
+-- Double + float is double if result does not fit float. \
+-- Float + float is float when no overflow. \
+-- Float + float is double when overflow. \
+-- Float + int is float when fits the float range. \
+-- Precision matters too. Double is used when need to avoid \
+-- precision loss. \
+tests = { \
+ {{'double', 1}, {'double', 1}, mp_float}, \
+ {{'double', flt_max}, {'double', flt_max}, mp_double}, \
+ {{'double', 1}, {'float', 1}, mp_float}, \
+ {{'double', flt_max}, {'float', flt_max}, mp_double}, \
+ {{'float', 1}, {'float', 1}, mp_float}, \
+ {{'float', flt_max}, {'float', flt_max}, mp_double}, \
+ {{'float', -flt_max}, {'float', -flt_max}, mp_double}, \
+ {{'float', 1}, {'int', 1}, mp_float}, \
+ {{'float', flt_max}, {'uint64_t', uint_max}, mp_double}, \
+ {{'float', 1.0001}, {'double', 1.0000000000001}, mp_double}, \
+}
+---
+...
+err = nil
+---
+...
+for i, test in pairs(tests) do \
+ local val1 = ffi.cast(test[1][1], test[1][2]) \
+ local val2 = ffi.cast(test[2][1], test[2][2]) \
+ local t = box.tuple.new({val1}) \
+ t = t:update({{'+', 1, val2}}) \
+ local m = msgpackffi.encode(t) \
+ if m:byte(1) ~= mp_array_1 or m:byte(2) ~= test[3] then \
+ err = {i, test, t, m:byte(1), m:byte(2)} \
+ break \
+ end \
+end
+---
+...
+err
+---
+- null
+...
diff --git a/test/box/update.test.lua b/test/box/update.test.lua
index c455bc1e1..724011c38 100644
--- a/test/box/update.test.lua
+++ b/test/box/update.test.lua
@@ -269,3 +269,48 @@ t:update({{'=', 3, map}})
s:update(1, {{'=', 3, map}})
s:drop()
+
+--
+-- gh-4701: arith operations should not truncate result when
+-- float + float fits double.
+--
+msgpackffi = require('msgpackffi')
+mp_array_1 = 0x91
+mp_double = 0xcb
+mp_float = 0xca
+flt_max = 3.402823466e+38
+uint_max = 18446744073709551615ULL
+tests = { \
+-- Double + double is float if result fits. \
+ {{'double', 1}, {'double', 1}, mp_float}, \
+-- Double + double is double if result does not fit float. \
+ {{'double', flt_max}, {'double', flt_max}, mp_double}, \
+-- Double + float is float if result fits. \
+ {{'double', 1}, {'float', 1}, mp_float}, \
+-- Double + float is double if result does not fit float. \
+ {{'double', flt_max}, {'float', flt_max}, mp_double}, \
+-- Float + float is float when no overflow. \
+ {{'float', 1}, {'float', 1}, mp_float}, \
+-- Float + float is double when overflow. \
+ {{'float', flt_max}, {'float', flt_max}, mp_double}, \
+ {{'float', -flt_max}, {'float', -flt_max}, mp_double}, \
+-- Float + int is float when fits the float range. \
+ {{'float', 1}, {'int', 1}, mp_float}, \
+ {{'float', flt_max}, {'uint64_t', uint_max}, mp_double}, \
+-- Precision matters too. Double is used when need to avoid \
+-- precision loss. \
+ {{'float', 1.0001}, {'double', 1.0000000000001}, mp_double}, \
+}
+err = nil
+for i, test in pairs(tests) do \
+ local val1 = ffi.cast(test[1][1], test[1][2]) \
+ local val2 = ffi.cast(test[2][1], test[2][2]) \
+ local t = box.tuple.new({val1}) \
+ t = t:update({{'+', 1, val2}}) \
+ local m = msgpackffi.encode(t) \
+ if m:byte(1) ~= mp_array_1 or m:byte(2) ~= test[3] then \
+ err = {i, test, t, m:byte(1), m:byte(2)} \
+ break \
+ end \
+end
+err
--
2.21.1 (Apple Git-122.3)
next reply other threads:[~2020-02-23 14:11 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-23 14:11 Vladislav Shpilevoy [this message]
2020-02-24 17:57 ` Kirill Yukhin
-- strict thread matches above, loose matches on Subject: below --
2020-01-26 16:44 Vladislav Shpilevoy
2020-01-26 18:45 ` Nikita Pettik
2020-01-27 21:56 ` Vladislav Shpilevoy
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=2b1afed1b05757c177b2fbb2efbf7643670c43cc.1582466958.git.v.shpilevoy@tarantool.org \
--to=v.shpilevoy@tarantool.org \
--cc=kyukhin@tarantool.org \
--cc=tarantool-patches@dev.tarantool.org \
--subject='Re: [Tarantool-patches] [PATCH 1/1] tuple: don'\''t truncate float in :update()' \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox