Messages in this thread | | | Subject | Re: New resources - pls, explain :-( | Date | Fri, 6 Aug 1999 19:50:24 +0100 (BST) | From | Alan Cox <> |
| |
> It says that I propose pci_request_device() allocating _the_resources_, so that > the problems we were speaking about some mails ago will be solved and usual > drivers still won't need to do any resource magic -- they'll just say > "grab this device and anything that comes with it".
I still need this anyway for I2O btw - I need to go from bus, slot, func to pci_dev then to grab its resources.
> But anyway, I feel that making pci_request_device() allocating all > the resources is a better solution.
It solves the problem for some devices, and makes them a lot cleaner in their resource handling. It certainly cleans up code and tidies up a lot of the potential for resource leaks in error cases.
Legacy resources need direct handling, special cases do too. No big deal
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |