daf73baeb9f1856ef42c176f7bff5d528baa457a
explain/proc.md
... | ... | @@ -27,9 +27,10 @@ There are a few details that are crucial for sizecoding stuff. On program entry: |
27 | 27 | ](https://pcy.ulyssis.be/intelrefspec.pdf). |
28 | 28 | Data on which SIMD instructions are working is sometimes stored on the stack. |
29 | 29 | *This means that, if you do not manually realign the stack, crashes will |
30 | - happen when doing SIMD. **This code may be in libraries you're depending on, |
|
30 | + happen when doing SIMD. __This code may be in libraries you're depending on, |
|
31 | 31 | and depending on the distro, libraries may or may not be compiled with SIMD |
32 | - instructions!** * This can be fixed with one byte: `push rax`. |
|
32 | + instructions!__* This can be fixed with one byte: `push rax`. |
|
33 | + * On `arm`, the stack seems to be [aligned to 8 bytes](https://wiki.debian.org/ArmEabiPort#Stack_alignment). |
|
33 | 34 | * Lots of interesting data is placed on the stack at program entry. See the |
34 | 35 | second lwn article for details. |
35 | 36 | * For dynamic linking-related stuff on program entry, see [this |