Messages in this thread | | | Date | Fri, 20 Jun 2008 11:47:12 -0400 (EDT) | From | Mikulas Patocka <> | Subject | Re: stack overflow on Sparc64 |
| |
> It could still be improved a lot, however. > >> Long-term consideration: Is it possible to implement interrupt stacks on >> sparc64? Functions on sparc eat stack much more aggressively than on other >> architectures (minimum stack size for a function is 192 bytes). > > I had a patch but at the time I wrote it (several years ago) I > couldn't make it stable enough to put mainline, I may resurrect it. > > I just did a quick scan and I can't find the last copy I had, and > things have changed enough that I'd probably work from scratch > anyways. > > But the level of recursion possible by the current device layer is > excessive and needs to be curtained irrespective of these generic > wakeup and sparc64 interrupt stack issues.
I took another few traces (to track the whole stack content) and there is another problem: nested interrupts. Does Sparc64 limit them somehow?
sys_call_table timer_interrupt irq_exit do_softirq __do_softirq run_timer_softirq _spin_unlock sys_call_table handler_irq handler_fasteoi_irq handle_irq_event ide_intr ide_dma_intr task_end_request ide_end_request __ide_end_request ...
Mikulas
| |