Commit graph

1360 commits

Author SHA1 Message Date
Riccardo Schirone
784aa48c05
Fix cmd_substitution in grep_specifier ##newshell (#16625)
* Allow cmd_substitution in grep_specifier and set core->tmpseek
2020-04-18 11:07:25 +02:00
Florian Märkl
f49659a4bc
Make rabin2 prefix Test more explicit for debugging #16452 (#16622) 2020-04-18 11:05:29 +02:00
Florian Märkl
dcf0c7049f
Chop Basic Blocks instead of re-analyzing in noreturn propagation (#16612) 2020-04-17 19:51:47 +02:00
Khairul Azhar Kasmiran
4afa4296ea
x86_32/64: Don't assemble overlarge MOV imm32 (#16619) 2020-04-17 22:36:58 +08:00
Florian Märkl
dae5bb43ee
Add a test for noreturn propagation (#16611) 2020-04-16 18:48:20 +02:00
Khairul Azhar Kasmiran
538563149b
x86_64: Fix MOV r32, -imm32 encoding (#16609) ##asm 2020-04-16 16:30:32 +02:00
Khairul Azhar Kasmiran
a275bad0db
x86_64: Fix REX.B for MOV r64-extended, imm32/64 (#16602) 2020-04-16 13:23:02 +08:00
Khairul Kasmiran
7b4bee8cff x86_64: Add some tests and remove some unused MOV C7 code 2020-04-14 23:48:47 +02:00
Khairul Azhar Kasmiran
a43831ba96 Fix #16433 - Use MOV opcode B8+ for MOV r64, <0x80000000 to 0xffffffff> #16572 ##asm
Detailed description

I've confirmed #16433 (comment) (including the nasm rax -> eax bug) and thus this pr
makes the x86_64 assembler use the MOV B8+ encoding (MOV r64, imm64) when the immediate
is between 0x80000000 and 0xffffffff for both:

Consistency with GNU assembler.  It actually does seem to be the expected encoding.
This does mean that mov eax, 0xffffffff and mov rax, 0xffffffff will have very different semantics.
2020-04-13 21:23:39 +02:00
pancake
346eb5bd38
Move test/new/* into test/ ##tests (#16511) 2020-04-13 19:49:23 +02:00