Timeline for Is IDA pulling my leg - or can REX.W sometimes not be determined in static analysis?
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 6, 2018 at 6:52 | history | tweeted | twitter.com/StackReverseEng/status/1015126335327330305 | ||
| Jul 5, 2018 at 21:19 | comment | added | Peter Cordes |
duplicate of stackoverflow Why do x86-64 instructions on 32-bit registers zero the upper part of the full 64-bit register?, and see also What is the best way to set a register to zero in x86 assembly: xor, mov or and? which explains that 32-bit operand-size is always best (e.g. xor r10d, r10d is not shorter than xor r10,r10, but Silvermont only recognizes 32-bit operand size as a zeroing idiom!)
|
|
| Jul 5, 2018 at 15:22 | vote | accept | 0xC0000022L♦ | ||
| Jul 5, 2018 at 14:31 | answer | added | jakobbotsch | timeline score: 15 | |
| Jul 5, 2018 at 14:24 | comment | added | NirIzr | Could you include some related code? say, code using E/RDX, additional code manipulating it prior to XOR instruction, etc. I've recently encountered a similar issue where after some more thorough investigation the upper half of RDX was indeed always zero, making this an unexploitable bug by the developer. | |
| Jul 5, 2018 at 14:16 | history | asked | 0xC0000022L♦ | CC BY-SA 4.0 |