Warn when coercing NAN to other types by Girgias · Pull Request #19573 · php/php-src

Girgias · 2025-08-24T13:34:24Z

RFC: https://wiki.php.net/rfc/warnings-php-8-5#coercing_nan_to_other_types

github-actions · 2025-08-29T18:10:25Z

AWS x86_64 (c7i.24xl)

Attribute	Value
Environment	aws
Runner	host
Instance type	c7i.metal-24xl (dedicated)
Architecture	x86_64
CPU	48 cores
CPU settings	disabled deeper C-states, disabled turbo boost, disabled hyper-threading
RAM	188 GB
Kernel	6.1.147-172.266.amzn2023.x86_64
OS	Amazon Linux 2023.8.20250818
GCC	14.2.1
Time	2025-08-29 17:20:09 UTC

Laravel 12.2.0 demo app - 100 consecutive runs, 50 warmups, 100 requests (sec)

PHP	Min	Max	Std dev	Rel std dev %	Mean	Mean diff %	Median	Median diff %	Skew	P-value	Instr count	Memory
PHP - baseline@fc46	0.46565	0.46818	0.00045	0.10%	0.46637	0.00%	0.46629	0.00%	1.515	0.999	176423932	44.01 MB
PHP - 8.5-warn-nan-coercion	0.45876	0.46679	0.00075	0.16%	0.46494	-0.31%	0.46499	-0.28%	-5.442	0.000	176485965	44.07 MB

Symfony 2.7.0 demo app - 100 consecutive runs, 50 warmups, 100 requests (sec)

PHP	Min	Max	Std dev	Rel std dev %	Mean	Mean diff %	Median	Median diff %	Skew	P-value	Instr count	Memory
PHP - baseline@fc46	0.72046	0.73259	0.00155	0.21%	0.72261	0.00%	0.72217	0.00%	3.109	0.999	287234974	40.16 MB
PHP - 8.5-warn-nan-coercion	0.72253	0.72854	0.00100	0.14%	0.72405	0.20%	0.72384	0.23%	1.831	0.000	287280133	40.53 MB

Wordpress 6.2 main page - 100 consecutive runs, 20 warmups, 20 requests (sec)

PHP	Min	Max	Std dev	Rel std dev %	Mean	Mean diff %	Median	Median diff %	Skew	P-value	Instr count	Memory
PHP - baseline@fc46	0.57464	0.58357	0.00228	0.39%	0.58060	0.00%	0.58151	0.00%	-1.225	0.999	1119139470	43.91 MB
PHP - 8.5-warn-nan-coercion	0.57682	0.58512	0.00218	0.37%	0.58209	0.26%	0.58298	0.25%	-1.361	0.000	1119813967	43.91 MB

bench.php - 100 consecutive runs, 10 warmups, 2 requests (sec)

PHP	Min	Max	Std dev	Rel std dev %	Mean	Mean diff %	Median	Median diff %	Skew	P-value	Instr count	Memory
PHP - baseline@fc46	0.42476	0.55444	0.02724	6.28%	0.43407	0.00%	0.42786	0.00%	4.179	0.999	2020744423	26.98 MB
PHP - 8.5-warn-nan-coercion	0.42534	0.58863	0.03210	7.34%	0.43703	0.68%	0.42765	-0.05%	3.477	0.582	2020694323	27.05 MB

arnaud-lb · 2025-08-29T18:48:32Z

+				 * Those optimizations are not safe if the other operand end up being NAN
+				 * as BOOL/BOOL_NOT will warn which IS_EQUAL/IS_NOT_EQUAL do not.


Triggered a benchmark to check the effect of this: valgrind shows no regression (changes are under 0.0x%).

I do wonder if this is not actually a problem for switch cases too now that I think of.

It would be worth it to check

Yeah, it seems I completely broke switch statements...

<?php $nan = fdiv(0, 0); switch ($nan) { case true: echo "true"; break; case false: echo "false"; break; } ?>

Returns nothing now, even without opcache :|

I wonder if this is important for canonicalization.

arnaud-lb · 2025-09-02T06:49:17Z

+				 * Those optimizations are not safe if the other operand end up being NAN
+				 * as BOOL/BOOL_NOT will warn which IS_EQUAL/IS_NOT_EQUAL do not.


It would be worth it to check

RFC: https://wiki.php.net/rfc/warnings-php-8-5#coercing_nan_to_other_types

iluuu1994

Couldn't see anything obviously incorrect.

iluuu1994 · 2025-09-22T14:21:50Z

+				 * Those optimizations are not safe if the other operand end up being NAN
+				 * as BOOL/BOOL_NOT will warn which IS_EQUAL/IS_NOT_EQUAL do not.


I wonder if this is important for canonicalization.

Co-authored-by: Ilija Tovilo <ilija.tovilo@me.com>

edorian

No RM objections 👍

TimWolla

Did look in particular detail. Not happy about the optimizations we lose 😔

IS_EQUAL does not warn, whereas BOOL/BOOL_NOT do now

staabm · 2025-09-24T13:54:27Z

was this change intentionally only implemented for NAN, but not for INF?

it seems INF has similar cast semantics: https://3v4l.org/lYIng

Girgias · 2025-09-24T15:55:33Z

was this change intentionally only implemented for NAN, but not for INF?

it seems INF has similar cast semantics: https://3v4l.org/lYIng

Yes, that's normal. NAN is meant to be "uncomparable" and only occurs in "nonsensical" FP operations, +-INF is just a very large value, however if you try to cast them to INT you would get another warning (done in a separate PR).

This PR was merged into the 6.4 branch. Discussion ---------- do not coerce NAN to other types | Q | A | ------------- | --- | Branch? | 6.4 | Bug fix? | no | New feature? | no | Deprecations? | no | Issues | | License | MIT see https://wiki.php.net/rfc/warnings-php-8-5#coercing_nan_to_other_types and php/php-src#19573 Commits ------- cd93939 do not coerce NAN to other types

Girgias added RFC Waiting on RM labels Aug 24, 2025

github-actions Bot added Category: Engine Category: Optimizer ABI break labels Aug 24, 2025

Girgias force-pushed the 8.5-warn-nan-coercion branch from d43ae9a to bfb238b Compare August 24, 2025 17:00

github-actions Bot added Extension: opcache Extension: standard labels Aug 24, 2025

Girgias force-pushed the 8.5-warn-nan-coercion branch 5 times, most recently from 8da693f to 729488f Compare August 29, 2025 15:06

Girgias requested a review from arnaud-lb August 29, 2025 15:38

arnaud-lb reviewed Aug 29, 2025

View reviewed changes

Girgias force-pushed the 8.5-warn-nan-coercion branch 2 times, most recently from 37c74ce to 7c29b7f Compare August 31, 2025 17:17

Girgias requested a review from arnaud-lb September 1, 2025 21:55

arnaud-lb reviewed Sep 2, 2025

View reviewed changes

Girgias force-pushed the 8.5-warn-nan-coercion branch from 7c29b7f to 895b751 Compare September 3, 2025 11:21

Girgias added this to the PHP 8.5 milestone Sep 13, 2025

Girgias force-pushed the 8.5-warn-nan-coercion branch 2 times, most recently from 6978b51 to 1640beb Compare September 22, 2025 01:01

Girgias added 2 commits September 22, 2025 03:04

core: Add new test for coercions from NAN

bbe346d

core: Warn when coercing NAN to other types

d82cfee

RFC: https://wiki.php.net/rfc/warnings-php-8-5#coercing_nan_to_other_types

Girgias force-pushed the 8.5-warn-nan-coercion branch from 1640beb to d82cfee Compare September 22, 2025 02:05

Girgias requested review from a team and arnaud-lb September 22, 2025 02:14

Girgias marked this pull request as ready for review September 22, 2025 08:41

Girgias requested review from bukka and dstogov as code owners September 22, 2025 08:41

Girgias requested a review from iluuu1994 September 22, 2025 10:07

[skip-ci] Remove whitespace changes

bb64442

iluuu1994 reviewed Sep 22, 2025

View reviewed changes

Girgias and others added 3 commits September 22, 2025 18:01

[skip ci] Comment nits

6e1d6ff

Co-authored-by: Ilija Tovilo <ilija.tovilo@me.com>

Address review comments

95ff6b7

Add some set_error_handler tests

5bc6fb1

edorian approved these changes Sep 22, 2025

View reviewed changes

edorian requested a review from TimWolla September 22, 2025 17:27

Precompute bool result before warning is emitted

0d7a7fc

TimWolla reviewed Sep 22, 2025

View reviewed changes

Comment thread Zend/zend_compile.c Outdated

Comment thread Zend/zend_operators.c

Girgias added 4 commits September 22, 2025 21:57

More tests and fixes for set_error_handler nonsense

941811f

Remove old compile time optimization

975aa39

IS_EQUAL does not warn, whereas BOOL/BOOL_NOT do now

Re-add one optimization in case we have a range (meaning not NAN)

cfa7c99

Make tests more robust

58d90b9

Girgias closed this in 320fe29 Sep 23, 2025

Girgias deleted the 8.5-warn-nan-coercion branch September 23, 2025 10:17

sebastianbergmann mentioned this pull request Sep 24, 2025

Warnings on PHP 8.5 sebastianbergmann/exporter#73

Closed

staabm mentioned this pull request Sep 24, 2025

Fix PHP8.5 Warning: "unexpected NAN value was coerced to string" phpstan/phpstan-src#4368

Merged

xabbuh mentioned this pull request Sep 25, 2025

do not coerce NAN to other types symfony/symfony#61844

Merged

jrfnl mentioned this pull request Oct 8, 2025

Account for changes introduced in PHP 8.5 PHPCompatibility/PHPCompatibility#1849

Open

70 tasks

ausi mentioned this pull request Sep 11, 2026

print_r() results in warning: unexpected NAN value was coerced to string #23662

Open

		* Those optimizations are not safe if the other operand end up being NAN
		* as BOOL/BOOL_NOT will warn which IS_EQUAL/IS_NOT_EQUAL do not.

Uh oh!

Conversation

Girgias commented Aug 24, 2025

Uh oh!

github-actions Bot commented Aug 29, 2025

AWS x86_64 (c7i.24xl)

Laravel 12.2.0 demo app - 100 consecutive runs, 50 warmups, 100 requests (sec)

Symfony 2.7.0 demo app - 100 consecutive runs, 50 warmups, 100 requests (sec)

Wordpress 6.2 main page - 100 consecutive runs, 20 warmups, 20 requests (sec)

bench.php - 100 consecutive runs, 10 warmups, 2 requests (sec)

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

arnaud-lb Aug 29, 2025

Choose a reason for hiding this comment

Uh oh!

Girgias Aug 31, 2025

Choose a reason for hiding this comment

Uh oh!

arnaud-lb Sep 2, 2025

Choose a reason for hiding this comment

Uh oh!

Girgias Sep 2, 2025

Choose a reason for hiding this comment

Uh oh!

iluuu1994 Sep 22, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

arnaud-lb Sep 2, 2025

Choose a reason for hiding this comment

Uh oh!

iluuu1994 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

iluuu1994 Sep 22, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

edorian left a comment

Choose a reason for hiding this comment

Uh oh!

TimWolla left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

staabm commented Sep 24, 2025

Uh oh!

Girgias commented Sep 24, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants