| Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 1 | # |
| 2 | # PCI configuration |
| 3 | # |
| 4 | config PCI_MSI |
| 5 | bool "Message Signaled Interrupts (MSI and MSI-X)" |
| 6 | depends on PCI |
| 7 | depends on (X86_LOCAL_APIC && X86_IO_APIC) || IA64 |
| 8 | help |
| 9 | This allows device drivers to enable MSI (Message Signaled |
| 10 | Interrupts). Message Signaled Interrupts enable a device to |
| 11 | generate an interrupt using an inbound Memory Write on its |
| 12 | PCI bus instead of asserting a device IRQ pin. |
| 13 | |
| Matthew Wilcox | 309e57d | 2006-03-05 22:33:34 -0700 | [diff] [blame] | 14 | Use of PCI MSI interrupts can be disabled at kernel boot time |
| 15 | by using the 'pci=nomsi' option. This disables MSI for the |
| 16 | entire system. |
| 17 | |
| Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 18 | If you don't know what to do here, say N. |
| 19 | |
| Greg Kroah-Hartman | 0f397f8 | 2006-07-18 10:59:59 -0700 | [diff] [blame] | 20 | config PCI_MULTITHREAD_PROBE |
| 21 | bool "PCI Multi-threaded probe (EXPERIMENTAL)" |
| Adrian Bunk | bb44c30 | 2006-10-27 16:12:30 -0700 | [diff] [blame^] | 22 | depends on PCI && EXPERIMENTAL && BROKEN |
| Greg Kroah-Hartman | 0f397f8 | 2006-07-18 10:59:59 -0700 | [diff] [blame] | 23 | help |
| 24 | Say Y here if you want the PCI core to spawn a new thread for |
| 25 | every PCI device that is probed. This can cause a huge |
| 26 | speedup in boot times on multiprocessor machines, and even a |
| 27 | smaller speedup on single processor machines. |
| 28 | |
| 29 | But it can also cause lots of bad things to happen. A number |
| 30 | of PCI drivers can not properly handle running in this way, |
| 31 | some will just not work properly at all, while others might |
| 32 | decide to blow up power supplies with a huge load all at once, |
| 33 | so use this option at your own risk. |
| 34 | |
| 35 | It is very unwise to use this option if you are not using a |
| 36 | boot process that can handle devices being created in any |
| 37 | order. A program that can create persistant block and network |
| 38 | device names (like udev) is a good idea if you wish to use |
| 39 | this option. |
| 40 | |
| 41 | Again, use this option at your own risk, you have been warned! |
| 42 | |
| 43 | When in doubt, say N. |
| 44 | |
| Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 45 | config PCI_DEBUG |
| 46 | bool "PCI Debugging" |
| 47 | depends on PCI && DEBUG_KERNEL |
| 48 | help |
| 49 | Say Y here if you want the PCI core to produce a bunch of debug |
| 50 | messages to the system log. Select this if you are having a |
| 51 | problem with PCI support and want to see more of what is going on. |
| 52 | |
| 53 | When in doubt, say N. |
| 54 | |
| Eric W. Biederman | 8b955b0 | 2006-10-04 02:16:55 -0700 | [diff] [blame] | 55 | config HT_IRQ |
| 56 | bool "Interrupts on hypertransport devices" |
| 57 | default y |
| Adrian Bunk | fbab41c | 2006-10-11 01:22:04 -0700 | [diff] [blame] | 58 | depends on PCI && X86_LOCAL_APIC && X86_IO_APIC |
| Eric W. Biederman | 8b955b0 | 2006-10-04 02:16:55 -0700 | [diff] [blame] | 59 | help |
| 60 | This allows native hypertransport devices to use interrupts. |
| 61 | |
| 62 | If unsure say Y. |