diff options
author | Grant Grundler <grundler@parisc-linux.org> | 2007-12-24 02:08:51 -0500 |
---|---|---|
committer | Greg Kroah-Hartman <gregkh@suse.de> | 2008-02-01 18:04:28 -0500 |
commit | d48b5d3a50c06357c721e81fa9354598282b6549 (patch) | |
tree | 21f90d5f2a088a7bad54dad34bde16d9ade90684 | |
parent | 7cbe5b6005f80de33a205d3052cdc89aacaac07c (diff) |
PCI: Remove pci_enable_device_bars() from documentation
Patch below removes pci_enable_device_bars() from Documentation/pci.txt .
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-rw-r--r-- | Documentation/pci.txt | 37 |
1 files changed, 1 insertions, 36 deletions
diff --git a/Documentation/pci.txt b/Documentation/pci.txt index 7754f5aea4e9..72b20c639596 100644 --- a/Documentation/pci.txt +++ b/Documentation/pci.txt | |||
@@ -274,8 +274,6 @@ the PCI device by calling pci_enable_device(). This will: | |||
274 | o allocate an IRQ (if BIOS did not). | 274 | o allocate an IRQ (if BIOS did not). |
275 | 275 | ||
276 | NOTE: pci_enable_device() can fail! Check the return value. | 276 | NOTE: pci_enable_device() can fail! Check the return value. |
277 | NOTE2: Also see pci_enable_device_bars() below. Drivers can | ||
278 | attempt to enable only a subset of BARs they need. | ||
279 | 277 | ||
280 | [ OS BUG: we don't check resource allocations before enabling those | 278 | [ OS BUG: we don't check resource allocations before enabling those |
281 | resources. The sequence would make more sense if we called | 279 | resources. The sequence would make more sense if we called |
@@ -605,40 +603,7 @@ device lists. This is still possible but discouraged. | |||
605 | 603 | ||
606 | 604 | ||
607 | 605 | ||
608 | 10. pci_enable_device_bars() and Legacy I/O Port space | 606 | 10. MMIO Space and "Write Posting" |
609 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | ||
610 | |||
611 | Large servers may not be able to provide I/O port resources to all PCI | ||
612 | devices. I/O Port space is only 64KB on Intel Architecture[1] and is | ||
613 | likely also fragmented since the I/O base register of PCI-to-PCI | ||
614 | bridge will usually be aligned to a 4KB boundary[2]. On such systems, | ||
615 | pci_enable_device() and pci_request_region() will fail when | ||
616 | attempting to enable I/O Port regions that don't have I/O Port | ||
617 | resources assigned. | ||
618 | |||
619 | Fortunately, many PCI devices which request I/O Port resources also | ||
620 | provide access to the same registers via MMIO BARs. These devices can | ||
621 | be handled without using I/O port space and the drivers typically | ||
622 | offer a CONFIG_ option to only use MMIO regions | ||
623 | (e.g. CONFIG_TULIP_MMIO). PCI devices typically provide I/O port | ||
624 | interface for legacy OSes and will work when I/O port resources are not | ||
625 | assigned. The "PCI Local Bus Specification Revision 3.0" discusses | ||
626 | this on p.44, "IMPLEMENTATION NOTE". | ||
627 | |||
628 | If your PCI device driver doesn't need I/O port resources assigned to | ||
629 | I/O Port BARs, you should use pci_enable_device_bars() instead of | ||
630 | pci_enable_device() in order not to enable I/O port regions for the | ||
631 | corresponding devices. In addition, you should use | ||
632 | pci_request_selected_regions() and pci_release_selected_regions() | ||
633 | instead of pci_request_regions()/pci_release_regions() in order not to | ||
634 | request/release I/O port regions for the corresponding devices. | ||
635 | |||
636 | [1] Some systems support 64KB I/O port space per PCI segment. | ||
637 | [2] Some PCI-to-PCI bridges support optional 1KB aligned I/O base. | ||
638 | |||
639 | |||
640 | |||
641 | 11. MMIO Space and "Write Posting" | ||
642 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | 607 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
643 | 608 | ||
644 | Converting a driver from using I/O Port space to using MMIO space | 609 | Converting a driver from using I/O Port space to using MMIO space |