Fix incorrect handling of ZEND_ACC_FINAL flag in JIT (#16778)

Fix GH-16393: Assertion failure in ext/opcache/jit/zend_jit.c:2897

Fix JIT support for static properties of self and parent

Use original op_array when JIT compiling a Closure

zend_jit() assumes that Closure op_arrays have no scope, but this is not true
when using the hot counters, first exec, or trace trigger

Use original op_array when JIT compiling a Closure

zend_jit() assumes that Closure op_arrays have no scope, but this is not true
when using the hot counters, first exec, or trace triggers as they use the
executed op_array, which is in case of Closures is a copy, with a scope.

In the tracing JIT this problem is avoided as we fetch the original op_array
when compiling a Closure. Here I replicate this for the hot counters and first
exec triggers.

Fixes GH-16186
Closes GH-16200

show more ...

JIT for INIT_STATIC_METHOD_CALL (#16206)

* JIT for INIT_STATIC_METHOD_CALL

* Fixed typo

* Fix missing LOAD

* Separate zend_get_known_class()

Implement JIT for ZEND_FETCH_STATIC_PROP_* and improve interpretation (#16157)

* Implement JIT for ZEND_FETCH_STATIC_PROP_* and improve interpretation

* Revert incorrect change

Implement JIT for ZEND_FETCH_STATIC_PROP_* and improve interpretation (#16157)

* Implement JIT for ZEND_FETCH_STATIC_PROP_* and improve interpretation

* Revert incorrect change

* Use FASTCALL calling convention

* Use EMPTY_SWITCH_DEFAULT_CASE

* Move the loading of the property info into zend_jit_uninit_static_prop()

show more ...

Add API to exempt function from being traced in JIT (#15559)

Internally accessible via zend_jit_blacklist_function / externally via opcache_jit_blacklist.
The functionality currently onl

Add API to exempt function from being traced in JIT (#15559)

Internally accessible via zend_jit_blacklist_function / externally via opcache_jit_blacklist.
The functionality currently only affects tracing JIT, but may be extended to other JIT modes in future.

show more ...

Fix merge into master

Interrupt while internal frame is on the stack (#14627)

* Check VM interrupt while internal frame is on top

* Use tab instead of spaces

* fix frame used in interrupt and re

Interrupt while internal frame is on the stack (#14627)

* Check VM interrupt while internal frame is on top

* Use tab instead of spaces

* fix frame used in interrupt and refactor

* remove unused failures for zend_jit_check_timeout

* Fix JIT support

Co-authored-by: Bob Weinand <bobwei9@hotmail.com>

* Fix the missing store to vm_interrupt

* Rename new functions

* Special case zend_interrupt_function in JIT code

* refactor to use ZEND_VM_SET_OPCODE_NO_INTERRUPT

* Split atomic exchange into load + store

It is difficult to determine performance of atomics sometimes. In this
case, the separate load+store is still correct, and a load does not
cause a modification, and might be faster for some platforms than an
exchange. A load+store is slower than an exchange, but we're fine
trading the penalty to the slow path and keeping the happy path faster.

---------

Co-authored-by: Bob Weinand <bobwei9@hotmail.com>

show more ...

Remove unintended ZEND_EXT_API usage (#15602)

[RFC] Convert exit (and die) from language constructs to functions (#13483)

RFC: https://wiki.php.net/rfc/exit-as-function

[RFC] Property hooks (#13455)

RFC: https://wiki.php.net/rfc/property-hooks

Co-authored-by: Nikita Popov <nikita.ppv@gmail.com>

Sync #if/ifdef/defined (#14508)

This syncs CPP macro conditions:
- _WIN32
- _WIN64
- HAVE_ALLOCA_H
- HAVE_ALPHASORT
- HAVE_ARPA_INET_H
- HAVE_CONFIG_H
- HAVE_DIRE

Sync #if/ifdef/defined (#14508)

This syncs CPP macro conditions:
- _WIN32
- _WIN64
- HAVE_ALLOCA_H
- HAVE_ALPHASORT
- HAVE_ARPA_INET_H
- HAVE_CONFIG_H
- HAVE_DIRENT_H
- HAVE_DLFCN_H
- HAVE_GETTIMEOFDAY
- HAVE_LIBDL
- HAVE_POLL_H
- HAVE_PWD_H
- HAVE_SCANDIR
- HAVE_SYS_FILE_H
- HAVE_SYS_PARAM_H
- HAVE_SYS_SOCKET_H
- HAVE_SYS_TIME_H
- HAVE_SYS_TYPES_H
- HAVE_SYS_WAIT_H
- HAVE_UNISTD_H
- PHP_WIN32
- ZEND_WIN32

These are either undefined or defined to 1 in Autotools and Windows.

Follow up of GH-5526 (-Wundef).

show more ...

JIT: Add CPU registers support for ASSIGN_OBJ and ASSIGN_OBJ_OP (#14303)

* JIT: Add CPU registers support for ASSIGN_OBJ and ASSIGN_OBJ_OP

* Fix tests failures

* Fix tests

JIT: Add CPU registers support for ASSIGN_OBJ and ASSIGN_OBJ_OP (#14303)

* JIT: Add CPU registers support for ASSIGN_OBJ and ASSIGN_OBJ_OP

* Fix tests failures

* Fix tests failures

* Add missing GUARD

show more ...

JIT: Add CPU registers support for FETCH_OBJ_R (#14253)

Omit FETCH_THIS in closures

Non-static closures are guaranteed to have $this. The existing comment
highlights this, but fails to handle it correctly.

Closes GH-14181

JIT: Improve x86_64 code generated for PHP Array Bounds Checks (#14246)

JIT: Implement CPU register usage for ASSIGN_DIM_OP (#14236)

JIT: Implement CPU register usage for ASSIGN_OP (#14235)

JIT: JIT: Implement CPU register usage for ISSET_ISEMPTY_DIM (#14227)

IT: Implement CPU register usage for FETCH_DIM_W (#14225)

JIT: Implement CPU register usage for ASSIGN_DIM (#14224)

JIT: Improve CPU register usage in FETCH_DIM_R, FETCH_DIM_IS and FETCH_LIST_R (#14222)

Fix GH-13834: Applying non-zero offset 36 to null pointer in zend_jit.c (#13846)

* Fix GH-13834: Applying non-zero offset 36 to null pointer in zend_jit.c

ssa_op can be NULL in func

Fix GH-13834: Applying non-zero offset 36 to null pointer in zend_jit.c (#13846)

* Fix GH-13834: Applying non-zero offset 36 to null pointer in zend_jit.c

ssa_op can be NULL in function JIT. Doing pointer arithmetic on a NULL
pointer is undefined behaviour. Undefined behaviour can be dangerous
because the optimizer may assume then that the variable is not actually
NULL.

To solve this:
1. Add ADVANCE_SSA_OP() to safely add an offset to ssa_op in zend_jit.c
2. For inference, add an extra offset argument to the helper functions.

To reproduce this, use Clang (not GCC) on a test like
sapi/cli/tests/gh12363.phpt (or other tests also work).

* Remove -fno-sanitize=pointer-overflow flag from CI

* Fix NULL pointer offsets added to the stack_map

* Fix an offset add on a potentially NULL ssa->ops

* Fix NULL pointer arithmetic in zend_range_info()

* Address review comments

show more ...

Implement stackless internal function calls

Co-authored-by: Dmitry Stogov <dmitry@zend.com>

Closes GH-12461