The implementation is using fixed size array allocated in asm module,
need to use proper array declaration for C source.
CID 1376405 - Memory corruption.
Details
Details
- Reviewers
cem - Commits
- rS320467: loader: chain load relocate data declaration is bad
Diff Detail
Diff Detail
- Repository
- rS FreeBSD src repository - subversion
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
| sys/boot/i386/libi386/libi386.h | ||
|---|---|---|
| 78 ↗ | (On Diff #29986) | If you explicitely specify array size in the extern declaration, you do not need relocater_size and can use nitems(). In revertse, if you use relocater_size, [] is enough in extern declaration. The later has the benefit of automatically catching updates. Why it is relocatEr and not relocator ? |
| sys/boot/i386/libi386/libi386.h | ||
|---|---|---|
| 78 ↗ | (On Diff #29986) | Yep, and in any case both will do for declaration. The E versus O, I actually do not mind either way really - this code was written quite some time ago, considering the variables below, I probably just did mix up;) Anyhow, I'll collect more feedback and then will likely have some update - the same bits are posted for illumos review. |
| sys/boot/i386/libi386/libi386.h | ||
|---|---|---|
| 77 ↗ | (On Diff #30184) | Should this still be an array, or just a singular struct or struct pointer? |
| sys/boot/i386/libi386/libi386.h | ||
|---|---|---|
| 77 ↗ | (On Diff #30184) | Ah right, I keep forgetting.. This feature actually has secondary consumer as well - but I haven't yet set the change/review up; the secondary consumer is the syslinux based software support - like freedos boot image with firmware update tools and things like that - and the extra slots are for it. If it is clear that we do not care about such functionality for freebsd, then it is reasonable to have just one structure instance. Otherwise the syslinux load needs all 3 slots. |