Support LONG MUL with overflow detection

Overflow detection for LONG MUL is added in this patch. Quite different
from 'subs' and 'adds' where overflow can be easily checked via the V

Support LONG MUL with overflow detection

Overflow detection for LONG MUL is added in this patch. Quite different
from 'subs' and 'adds' where overflow can be easily checked via the V
flags, LONG MUL wouldn't set the flags.

We use 'smulh' instruction to get the upper 64 bits of the 128-bit
result and check the top 65 bits to tell whether integer overflow
occurs. [1]

Note that LONG MUL can be substituted by 'adds' or 'lsl' in some cases.
Hence, flag 'use_mul' is introduced in order to select the proper
overflow check check instruction afterwards.

[1]
https://community.arm.com/developer/ip-products/processors/b/processors-ip-blog/posts/detecting-overflow-from-mul

Change-Id: I67e8287e9044c2a96b188d4bf6674736713abfe9

show more ...

Support for ASSIGN_DIM_OP

Support for ASSIGN_OP

Support for ISSET_DIM

Accurate RETVAL register usage

Fixed copy/paste error

Fixed incorrect register usage

Don't use FPR1 for consistency with x86

Support for FETCH_DIM* and ASSIGN_DIM (exception handling is incomplete)

Support for IS_IDENTICAL/IS_NOT_IDENTICAL

Fixed copy/paste error

Support for BOOL/BOOL_NOT with DOUBLE operand

Support for comparison opcodes

Support for CONCAT

Support for INC/DEC

Fixed possible incorrect assumption if constant is a power of 2.

Remove dead conditions

Support for LONG math (except of MUL overflow detection)

Revisit the encoding of immediate

Immediates in different lengths are used by AArch64 instructions.
Previous use of MAX_IMM12 is not accurate.

Note that immediate value can be o

Revisit the encoding of immediate

Immediates in different lengths are used by AArch64 instructions.
Previous use of MAX_IMM12 is not accurate.

Note that immediate value can be optionally shifted and this mode is not
supported currently. We may want to support this in the near future.

show more ...

Remove the duplicate macro DOUBLE_GET_ZVAL_DVAL

We have defined macro GET_ZVAL_DVAL to replace the SSE_GET_ZVAL_DVAL in
x86 implementation. Hence, macro DOUBLE_GET_ZVAL_DVAL is not neede

Remove the duplicate macro DOUBLE_GET_ZVAL_DVAL

We have defined macro GET_ZVAL_DVAL to replace the SSE_GET_ZVAL_DVAL in
x86 implementation. Hence, macro DOUBLE_GET_ZVAL_DVAL is not needed.

show more ...

Updates for commits between 121a0f7 and 12dcf34

1. Pre-allocated bytes are missing in function
zend_jit_assign_const_stub().

2. 'w' register should be used for macro SET_ZVAL_TY

Updates for commits between 121a0f7 and 12dcf34

1. Pre-allocated bytes are missing in function
zend_jit_assign_const_stub().

2. 'w' register should be used for macro SET_ZVAL_TYPE_INFO.

3. 'w' register should be used to load "num_args" in function
zend_jit_do_fcall().

4. Remove the local path name in test case recv_002.phpt

5. One option is disabled temporarily in [1] and several test cases
would fail, e.g. shift_right_003.phpt. I suppose new execution paths are
touched. We will support them in the near future.

[1] https://github.com/php/php-src/commit/63d673d

show more ...

Support for DOUBLE math

Support for most cases of BOOL/BOOL_NOT/JMPZNZ/JMP[N]Z_EX

Support for passing extra arguments.
Fixed ext/opcache/tests/jit/recv_001.phpt

Support for RECV/RECV_INIT