| -!- Jia` is now known as Jia | 10:03 | |
| derRichard | hmm, uclibs libpthread is really nasty. especially in the openrisc case where you have no cmpxchg | 13:00 |
|---|---|---|
| derRichard | this while (testandset(spinlock)) stuff in __pthread_acquire is utter crap. | 13:00 |
| juliusb | derRichard: what's up with the cmpxchg stuff? dose that usually rely on hardware semaphore support? | 14:02 |
| derRichard | juliusb: if an arch does not support a cmpxchg like instruction you have to go into the kernel to do locking. all this sched_yield(), nanosleep() stuff what ulibc (and dietlibc) does in this case is crap | 14:04 |
| derRichard | AFAIK arm has a cmpxchg system call for this. glibc uses it if no cmpxchg is available | 14:04 |
| juliusb | that's probably a good suggestion | 14:10 |
| juliusb | ARM does have a swap instruction, though, but it may not be usable in user mode | 14:11 |
| derRichard | juliusb: not all arm socs support it | 14:11 |
| derRichard | bbl, lunch | 14:11 |
| juliusb | well, they don't have it on the new stuff anyway (they prefer the exclusive accesses) | 14:11 |
| juliusb | but it sounds like a good idea to go into the kernel for it | 14:12 |
| derRichard | juliusb: going into the kernel is the only way to fix the issue. glibc does so for very good reasons | 14:49 |
| derRichard | i guess dielibc and ulibc never noticed the problem because you need more then a trivial test case to spot the bug | 14:50 |
| stekern | ah, I think I've found out what made clang/llvm-compiled linux bug out yesterday | 18:37 |
| derRichard | tell | 18:37 |
| stekern | looks like I've made functions that don't need a frame-pointer threat the fp (r2) with too little respect | 18:38 |
| stekern | i.e. it uses it as a temp-reg but doesn't save/restore it | 18:38 |
| stekern | fixing that made it progress further, but it's still crashing | 20:40 |
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!